INDEX

Explanations

phrases comparing or describing similarities between different things

oai_token-act-pair · gpt-3.5-turbo

discourse markers that introduce comparisons or similarities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Ł

-0.78

emen

-0.76

¿

-0.76

³

-0.74

ayne

-0.71

ewater

-0.71

»Ĵ

-0.70

uador

-0.70

ew

-0.70

eman

-0.70

POSITIVE LOGITS

 however

0.69

0.59

 inher

0.57

 except

0.57

 unfortunately

0.57

 opio

0.56

 ours

0.56

 though

0.56

 Manz

0.55

Activations Density 0.153%

phrases comparing or describing similarities between different things

discourse markers that introduce comparisons or similarities

No Comments

No Known Activations