INDEX

Explanations

them here looking for tokens indicating negative sentiment or disagreement

oai_token-act-pair · gpt-3.5-turbo

the phrase "No" as a repeated expression of refusal or negation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

equ

-0.68

 wagon

-0.67

 lingu

-0.66

 related

-0.63

apan

-0.63

 accounting

-0.62

 barg

-0.62

 particularly

-0.61

 exempl

-0.60

 around

-0.59

POSITIVE LOGITS

No

3.15

no

2.14

No

2.08

Yes

1.99

NO

1.72

Nothing

1.56

Never

1.53

Neither

1.52

Nobody

1.52

Not

1.45

Activations Density 0.015%

them here looking for tokens indicating negative sentiment or disagreement

the phrase "No" as a repeated expression of refusal or negation

No Comments

No Known Activations