INDEX

Explanations

words related to decision-making and justification

oai_token-act-pair · gpt-3.5-turbo

phrases related to foundational principles or bases for actions and beliefs

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Bye

-0.60

 Scrib

-0.58

 Maxwell

-0.55

osis

-0.55

:(

-0.55

 broke

-0.55

etts

-0.54

 informing

-0.54

 itch

-0.53

 testify

-0.53

POSITIVE LOGITS

 oneself

1.10

 yourselves

0.95

amera

0.88

illac

0.84

 themselves

0.84

senal

0.84

 blame

0.82

 herself

0.82

 Yourself

0.82

 yourself

0.80

Activations Density 0.378%

words related to decision-making and justification

phrases related to foundational principles or bases for actions and beliefs

No Comments

No Known Activations