INDEX

Explanations

situations or statements with negative connotations or outcomes

oai_token-act-pair · gpt-3.5-turbo

instances of the word "worse" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ettes

-0.78

ivating

-0.76

trl

-0.75

oir

-0.75

ools

-0.74

encers

-0.74

ieu

-0.73

arya

-0.71

ivo

-0.71

 Channel

-0.70

POSITIVE LOGITS

 than

1.04

 behaved

1.02

 Than

0.89

 nightmares

0.86

 catast

0.85

 consequences

0.79

than

0.77

 nightmare

0.76

nces

0.75

 scenarios

0.73

Activations Density 0.025%

situations or statements with negative connotations or outcomes

instances of the word "worse" in various contexts

No Comments

No Known Activations