INDEX

Explanations

topics related to brief information summaries

oai_token-act-pair · gpt-3.5-turbo

repetitions of the word "Brief"

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

artifacts

-0.86

CE

-0.77

OURCE

-0.72

pez

-0.70

GO

-0.69

ç¥ŀ

-0.67

SEA

-0.66

tackle

-0.65

ceans

-0.65

OF

-0.64

POSITIVE LOGITS

cases

1.13

case

1.07

ing

1.05

ly

0.93

 Brief

0.88

itude

0.87

rief

0.83

ings

0.80

gements

0.80

edIn

0.77

Activations Density 0.008%

topics related to brief information summaries

repetitions of the word "Brief"

No Comments

No Known Activations