INDEX

Explanations

specific examples

oai_token-act-pair · gpt-3.5-turbo

instances and examples used in explanations or discussions

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

inated

-0.68

YING

-0.66

sis

-0.65

ggles

-0.65

isters

-0.64

ocratic

-0.63

organic

-0.63

atures

-0.63

asca

-0.61

aments

-0.60

POSITIVE LOGITS

hesda

0.79

tainment

0.77

lihood

0.77

"@

0.76

forth

0.73

mma

0.70

Newsletter

0.69

 Kimmel

0.67

wagon

0.67

ðĿ

0.65

Activations Density 0.013%

specific examples

instances and examples used in explanations or discussions

No Comments

No Known Activations