INDEX

Explanations

phrases related to deception or dishonesty

oai_token-act-pair · gpt-3.5-turbo

instances of the word "had" indicating past actions or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

porting

-0.64

bie

-0.62

 Voters

-0.61

yles

-0.60

hack

-0.60

hammer

-0.59

bies

-0.58

umping

-0.58

 Recap

-0.58

FW

-0.58

POSITIVE LOGITS

 been

1.34

 undergone

1.17

 begun

1.05

 gone

1.04

been

0.99

 gotten

0.99

iths

0.98

 become

0.93

 taken

0.91

 come

0.90

Activations Density 0.152%

phrases related to deception or dishonesty

instances of the word "had" indicating past actions or events

No Comments

No Known Activations