INDEX

Explanations

mention of emergency response situations or individuals

oai_token-act-pair · gpt-3.5-turbo

references to the term "response" or related phrases indicating a reaction or action taken

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

ãĥ³

-0.63

âĶĢâĶĢ

-0.63

cery

-0.63

etheus

-0.62

rily

-0.62

urities

-0.62

 stirring

-0.62

akespeare

-0.61

ducers

-0.60

ruction

-0.60

POSITIVE LOGITS

 Resp

2.29

McF

0.81

pring

0.79

iosis

0.72

 Synd

0.65

 Bapt

0.64

Sym

0.63

 footwear

0.62

 Shoes

0.62

 Riding

0.62

Activations Density 0.004%

mention of emergency response situations or individuals

references to the term "response" or related phrases indicating a reaction or action taken

No Comments

No Known Activations