INDEX

Explanations

terms related to numerical values and percentages

oai_token-act-pair · gpt-3.5-turbo

numeric data and references to statistics or measurable quantities

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

autions

-0.78

linger

-0.76

ãĥĻ

-0.73

pret

-0.72

ãģĨ

-0.72

 teasp

-0.72

omen

-0.66

apter

-0.66

escription

-0.66

ETA

-0.66

POSITIVE LOGITS

NYC

0.58

Cyn

0.57

NSW

0.56

 Newtown

0.54

DOI

0.52

kWh

0.52

 Bots

0.52

LW

0.52

 Organic

0.51

 multi

0.51

Activations Density 1.564%

terms related to numerical values and percentages

numeric data and references to statistics or measurable quantities

No Comments

No Known Activations