INDEX

Explanations

phrases or words related to giving or receiving directions

oai_token-act-pair · gpt-3.5-turbo

references to guidance or navigation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

bon

-0.76

tle

-0.73

IGH

-0.72

Mi

-0.68

ony

-0.68

ighth

-0.66

osphere

-0.65

nia

-0.63

ggies

-0.62

Assembly

-0.62

POSITIVE LOGITS

 directions

1.54

 Directions

1.03

 direction

0.92

 instructions

0.82

pread

0.78

 thereto

0.75

autions

0.72

srf

0.70

 commands

0.69

icular

0.68

Activations Density 0.009%

phrases or words related to giving or receiving directions

references to guidance or navigation

No Comments

No Known Activations