INDEX

Explanations

code elements or syntax in a computer programming language, specifically related to GNU C++ syntax

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names associated with individuals or entities

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 timelines

-0.73

ĪĴ

-0.68

ilion

-0.66

ctica

-0.65

 entrants

-0.63

 Archdemon

-0.63

cffffcc

-0.63

 adversaries

-0.62

 Aval

-0.62

 judges

-0.62

POSITIVE LOGITS

mann

1.10

iegel

1.07

enstein

0.97

man

0.89

stein

0.88

geist

0.87

baum

0.86

berg

0.83

bach

0.82

stadt

0.82

Activations Density 0.011%

code elements or syntax in a computer programming language, specifically related to GNU C++ syntax

proper nouns, particularly names associated with individuals or entities

No Comments

No Known Activations