INDEX

Explanations

superlative rankings or positions

oai_token-act-pair · gpt-3.5-turbo

references to ranked lists or standings

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

ufact

-0.69

 Consent

-0.65

ajor

-0.64

cci

-0.62

Äĩ

-0.62

ool

-0.61

ustom

-0.61

 Advent

-0.60

asca

-0.60

 Lauder

-0.59

POSITIVE LOGITS

ographical

1.14

most

1.09

 tier

1.00

ography

0.98

ographically

0.98

 scorer

0.95

deck

0.94

 contenders

0.91

ographic

0.90

 drawer

0.89

Activations Density 0.035%

superlative rankings or positions

references to ranked lists or standings

No Comments

No Known Activations