INDEX

Explanations

phrases related to rates or measurements

oai_token-act-pair · gpt-3.5-turbo

mentions of statistical rates or measurements related to various topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

Ô

-0.80

KK

-0.77

 Cosponsors

-0.76

ctuary

-0.76

ittees

-0.71

 istg

-0.70

jad

-0.69

few

-0.68

ulla

-0.66

arna

-0.66

POSITIVE LOGITS

 course

0.80

 interest

0.80

efficiency

0.76

 measurement

0.75

 maturity

0.75

 parity

0.74

 income

0.71

 importance

0.71

 sorts

0.70

 allowable

0.70

Activations Density 0.402%

phrases related to rates or measurements

mentions of statistical rates or measurements related to various topics

No Comments

No Known Activations