INDEX

Explanations

references to significant decisions or actions being taken

oai_token-act-pair · gpt-3.5-turbo

instances of the word "move."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

Condition

-0.74

omial

-0.72

IZE

-0.61

 Curve

-0.59

iciency

-0.59

icum

-0.58

raid

-0.58

oola

-0.58

 concess

-0.58

 Koran

-0.57

POSITIVE LOGITS

 toward

1.11

 towards

1.10

backs

0.87

able

0.86

 forward

0.83

 away

0.82

over

0.81

overs

0.80

rers

0.79

iton

0.78

Activations Density 0.031%

references to significant decisions or actions being taken

instances of the word "move."

No Comments

No Known Activations