INDEX

Explanations

phrases related to small or minimal entities or activities

oai_token-act-pair · gpt-3.5-turbo

keywords related to "micro" concepts and phenomena

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

Wem

-0.69

 Moreno

-0.68

ĪĴ

-0.65

ktop

-0.64

 Sham

-0.63

Wol

-0.63

 Zimmer

-0.62

 Petty

-0.62

 Pole

-0.61

 Hutchinson

-0.61

POSITIVE LOGITS

batch

0.77

abytes

0.76

oute

0.75

abulary

0.74

Î³

0.74

lections

0.74

rules

0.72

kat

0.72

tiny

0.72

seconds

0.72

Activations Density 0.077%

phrases related to small or minimal entities or activities

keywords related to "micro" concepts and phenomena

No Comments

No Known Activations