INDEX

Explanations

mentions of disputes or conflicts involving various groups or individuals

oai_token-act-pair · gpt-3.5-turbo

punctuated lists or enumerations within sentences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

Marco

-0.63

 Slash

-0.62

STON

-0.61

Actor

-0.61

ãĥ¥

-0.59

 Heist

-0.59

(>

-0.58

(-

-0.58

 highs

-0.57

clusive

-0.56

POSITIVE LOGITS

are

0.94

 deserve

0.90

 constitute

0.89

 undertook

0.88

 have

0.88

 refuse

0.87

 chose

0.86

 intend

0.85

 were

0.85

 perceive

0.85

Activations Density 0.185%

mentions of disputes or conflicts involving various groups or individuals

punctuated lists or enumerations within sentences

No Comments

No Known Activations