INDEX

Explanations

assertive statements about different topics

oai_token-act-pair · gpt-3.5-turbo

phrases beginning with the article 'a,' typically indicating various concepts or classifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

Edit

-0.82

istar

-0.75

enery

-0.74

edit

-0.72

 edits

-0.72

 Emails

-0.70

Versions

-0.69

andon

-0.69

votes

-0.68

coins

-0.68

POSITIVE LOGITS

 cornerstone

1.11

 continuum

1.02

 rite

1.00

 prerequisite

0.97

 lucrative

0.94

 complex

0.92

 multif

0.92

 fascinating

0.92

 unique

0.91

 hallmark

0.91

Activations Density 0.220%

assertive statements about different topics

phrases beginning with the article 'a,' typically indicating various concepts or classifications

No Comments

No Known Activations