INDEX

Explanations

actions or tasks that need to be done or achieved

oai_token-act-pair · gpt-3.5-turbo

phrases related to processes and actions

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 cameo

-0.63

 à¨

-0.59

¿½

-0.58

è¦

-0.57

BuyableInstoreAndOnline

-0.57

 href

-0.55

 railing

-0.54

ighed

-0.53

ãĤ¸

-0.53

POSITIVE LOGITS

 safely

1.39

 properly

1.28

 efficiently

1.25

 correctly

1.10

 effectively

0.96

 smoothly

0.93

 securely

0.92

 accurately

0.91

 anew

0.89

Activations Density 0.677%

actions or tasks that need to be done or achieved

phrases related to processes and actions

No Comments

No Known Activations