INDEX

Explanations

phrases describing specific types or instances of items or concepts

oai_token-act-pair · gpt-3.5-turbo

phrases that introduce examples or instances

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

ombat

-0.73

ribution

-0.72

antage

-0.66

dollar

-0.66

orem

-0.65

ushima

-0.63

emi

-0.63

iliate

-0.62

Cause

-0.60

 Bore

-0.60

POSITIVE LOGITS

ties

0.79

cond

0.74

things

0.70

Osw

0.66

odon

0.65

types

0.61

requ

0.61

 embodiments

0.61

necess

0.60

 prec

0.60

Activations Density 0.030%

phrases describing specific types or instances of items or concepts

phrases that introduce examples or instances

No Comments

No Known Activations