INDEX

Explanations

phrases related to causality or conditions

oai_token-act-pair · gpt-3.5-turbo

punctuation marks, specifically commas

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

Redd

-0.79

çļ

-0.78

bryce

-0.74

-0.72

à¨

-0.67

ND

-0.66

redd

-0.65

papers

-0.64

Detailed

-0.63

fires

-0.63

POSITIVE LOGITS

aten

0.74

 barring

0.70

 unless

0.69

yne

0.64

ativity

0.62

atum

0.62

uh

0.62

amen

0.62

anos

0.62

pheus

0.62

Activations Density 0.059%

phrases related to causality or conditions

punctuation marks, specifically commas

No Comments

No Known Activations