INDEX

Explanations

phrases related to personal or professional achievements and attributes

oai_token-act-pair · gpt-3.5-turbo

key terms related to societal issues and change

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

arton

-0.59

awaru

-0.57

umption

-0.56

asions

-0.54

cies

-0.54

ancies

-0.54

mentioned

-0.50

istries

-0.50

Wem

-0.49

nia

-0.49

POSITIVE LOGITS

 unto

0.66

 fodder

0.63

 centerpiece

0.57

 underdog

0.56

 sleeper

0.55

 incarn

0.54

 pinnacle

0.54

 starter

0.53

 conduit

0.52

ender

0.52

Activations Density 1.008%

phrases related to personal or professional achievements and attributes

key terms related to societal issues and change

No Comments

No Known Activations