INDEX

Explanations

phrases related to conflicts or disagreements

oai_token-act-pair · gpt-3.5-turbo

the presence of conjunctions and phrases indicating comparisons or connections between ideas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Profile

-0.61

 Guerrero

-0.59

 Starr

-0.57

Sep

-0.57

 Pledge

-0.57

during

-0.56

successfully

-0.55

QB

-0.55

essage

-0.55

GD

-0.55

POSITIVE LOGITS

 wiser

0.71

 equals

0.68

eh

0.64

 shalt

0.62

 stroke

0.62

 pudding

0.61

 nutshell

0.61

 caveat

0.60

 caveats

0.60

pired

0.59

Activations Density 0.857%

phrases related to conflicts or disagreements

the presence of conjunctions and phrases indicating comparisons or connections between ideas

No Comments

No Known Activations