INDEX

Explanations

prepositions followed by specific locations or contexts

oai_token-act-pair · gpt-3.5-turbo

references to political events or situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

eret

-0.73

Reviewed

-0.68

imar

-0.67

bsite

-0.66

íķ

-0.64

IAS

-0.63

Fab

-0.62

 closet

-0.62

 derog

-0.61

 perspect

-0.61

POSITIVE LOGITS

marks

0.80

keeping

0.75

score

0.67

sburgh

0.66

tery

0.63

EMP

0.59

alion

0.58

okane

0.58

 Corm

0.57

mark

0.57

Activations Density 0.000%

prepositions followed by specific locations or contexts

references to political events or situations

No Comments

No Known Activations