INDEX

Explanations

phrases related to strong performances, particularly in the context of various events or fields

oai_token-act-pair · gpt-3.5-turbo

references to performance evaluations

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

vy

-0.83

yg

-0.80

yu

-0.73

cot

-0.71

sonian

-0.69

ski

-0.69

cott

-0.68

oslov

-0.68

sight

-0.68

ship

-0.67

POSITIVE LOGITS

 enhancement

0.87

imony

0.86

 evaluations

0.86

 appraisal

0.82

 enhancing

0.79

 performances

0.79

istics

0.78

enh

0.76

 evaluation

0.76

 performance

0.75

Activations Density 0.023%

phrases related to strong performances, particularly in the context of various events or fields

references to performance evaluations

No Comments

No Known Activations