INDEX

Explanations

phrases related to research findings or experimental results

oai_token-act-pair · gpt-3.5-turbo

references to experimental or survey outcomes

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

cer

-0.69

capital

-0.65

eways

-0.64

 hopping

-0.64

kers

-0.63

oos

-0.63

ACP

-0.62

vid

-0.62

 leisure

-0.61

Rim

-0.61

POSITIVE LOGITS

 results

1.13

results

1.10

Results

1.01

 result

0.93

 Results

0.92

result

0.85

iments

0.83

ĸļ

0.76

ULTS

0.76

inct

0.74

Activations Density 0.017%

phrases related to research findings or experimental results

references to experimental or survey outcomes

No Comments

No Known Activations