INDEX

Explanations

phrases related to learning from experiences or mistakes

oai_token-act-pair · gpt-3.5-turbo

references to learning from past experiences or mistakes

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

mouth

-0.88

inence

-0.86

å¥³

-0.85

imo

-0.83

isphere

-0.78

ometer

-0.73

olson

-0.73

puter

-0.73

cone

-0.72

âĹ¼

-0.72

POSITIVE LOGITS

 afar

1.01

 scratch

0.89

 experience

0.88

 whence

0.87

 Experience

0.84

 mistakes

0.82

 lessons

0.82

 Fukushima

0.81

 Lessons

0.79

 abroad

0.75

Activations Density 0.068%

phrases related to learning from experiences or mistakes

references to learning from past experiences or mistakes

No Comments

No Known Activations