INDEX

Explanations

words related to actions or commands directed towards a specific person or group

oai_token-act-pair · gpt-3.5-turbo

verbs indicating action or commands

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

mith

-0.64

SPONSORED

-0.63

printf

-0.58

 constitu

-0.57

 Mehran

-0.56

dfx

-0.56

]=

-0.55

 disg

-0.54

 spons

-0.53

cv

-0.53

POSITIVE LOGITS

 Yourself

1.39

 yourself

1.24

ments

1.16

 Your

1.15

 your

1.14

 yourselves

1.14

ings

1.07

ment

0.97

able

0.96

ables

0.95

Activations Density 0.301%

words related to actions or commands directed towards a specific person or group

verbs indicating action or commands

No Comments

No Known Activations