INDEX

Explanations

words related to malicious activities or harm

oai_token-act-pair · gpt-3.5-turbo

words related to fishing and the act of catching fish

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

ilogy

-0.75

abet

-0.68

 Aerospace

-0.67

 notebook

-0.65

psey

-0.64

clerosis

-0.64

supp

-0.63

ETF

-0.62

ensis

-0.62

xus

-0.62

POSITIVE LOGITS

ishes

0.97

ishing

0.93

ISH

0.83

Hung

0.77

ished

0.76

NESS

0.76

ãĤ¤ãĥĪ

0.74

ishment

0.72

 Vessel

0.70

ishly

0.68

Activations Density 0.008%

words related to malicious activities or harm

words related to fishing and the act of catching fish

No Comments

No Known Activations