INDEX

Explanations

numerical values preceded by a dollar sign

oai_token-act-pair · gpt-3.5-turbo

expressions that indicate actions or user engagement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

],"

-0.60

-0.57

"—

-0.56

existent

-0.56

querade

-0.56

 Mehran

-0.55

Ö

-0.54

foundland

-0.53

\.

-0.52

',"

-0.52

POSITIVE LOGITS

cknowled

0.68

cing

0.57

 furthermore

0.54

sequently

0.54

 Leaf

0.52

entimes

0.52

 latter

0.51

quartered

0.51

 drawback

0.50

 Result

0.50

Activations Density 0.762%

numerical values preceded by a dollar sign

expressions that indicate actions or user engagement

No Comments

No Known Activations