INDEX

Explanations

words related to comparative evaluation, specifically when something is judged to be somewhat less or more of a particular quality

oai_token-act-pair · gpt-3.5-turbo

the word "somewhat" and its various contexts and implications

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

abad

-0.77

ults

-0.77

iens

-0.74

aring

-0.73

eret

-0.72

elsen

-0.72

ULTS

-0.72

olan

-0.70

yrs

-0.70

yers

-0.68

POSITIVE LOGITS

 surprising

0.84

 unusual

0.83

 amusing

0.83

 cryptic

0.82

 obscure

0.82

 inaccurate

0.81

unn

0.80

 insensitive

0.80

 constrained

0.80

 confusing

0.79

Activations Density 0.016%

words related to comparative evaluation, specifically when something is judged to be somewhat less or more of a particular quality

the word "somewhat" and its various contexts and implications

No Comments

No Known Activations