INDEX

Explanations

words related to convenience or positive qualities of objects or ideas

oai_token-act-pair · gpt-3.5-turbo

adjectives and descriptors that convey positive characteristics or qualities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 worshipped

-0.76

eds

-0.74

usercontent

-0.73

$$$$

-0.71

everyone

-0.70

payers

-0.70

 depended

-0.70

hers

-0.69

hes

-0.69

asters

-0.68

POSITIVE LOGITS

 anecdote

1.04

tid

1.02

 infographic

1.01

 tale

0.99

 twist

0.94

 combination

0.94

 irony

0.93

 little

0.93

 illustration

0.92

 piece

0.92

Activations Density 0.155%

words related to convenience or positive qualities of objects or ideas

adjectives and descriptors that convey positive characteristics or qualities

No Comments

No Known Activations