INDEX

Explanations

references to railway lines

oai_token-act-pair · gpt-3.5-turbo

terms related to railways and transportation infrastructure

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

VIDIA

-0.79

uthor

-0.74

Beat

-0.73

Brew

-0.70

Interview

-0.69

hemy

-0.68

untu

-0.67

rolet

-0.67

 prose

-0.66

Bloom

-0.66

POSITIVE LOGITS

roads

1.08

route

0.89

 route

0.88

 junction

0.86

 perpendicular

0.86

runners

0.81

 crossings

0.79

 routes

0.79

runner

0.78

course

0.78

Activations Density 0.058%

references to railway lines

terms related to railways and transportation infrastructure

No Comments

No Known Activations