INDEX

Explanations

comparative statements indicating a difference in quantity or degree

oai_token-act-pair · gpt-3.5-turbo

comparative phrases indicating lower quantities or values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

oiler

-0.69

ointed

-0.63

ongo

-0.62

oint

-0.62

 Everywhere

-0.61

 distinction

-0.60

allion

-0.59

oke

-0.57

escal

-0.57

eah

-0.57

POSITIVE LOGITS

 usual

1.07

ptions

0.91

 theirs

0.84

 typical

0.81

usual

0.80

 normal

0.79

 expected

0.78

 dictated

0.76

 ordinary

0.76

average

0.75

Activations Density 0.142%

comparative statements indicating a difference in quantity or degree

comparative phrases indicating lower quantities or values

No Comments

No Known Activations