INDEX

Explanations

references to physical attacks or conflicts

oai_token-act-pair · gpt-3.5-turbo

references to mugging incidents and associated discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

cells

-0.75

cel

-0.75

ed

-0.75

ters

-0.74

coe

-0.73

nd

-0.72

ducers

-0.72

ek

-0.71

er

-0.69

seed

-0.69

POSITIVE LOGITS

ifully

0.90

iless

0.80

iful

0.78

 Hots

0.77

enance

0.74

istics

0.70

lust

0.67

llan

0.65

ãĢĳ

0.65

ainment

0.65

Activations Density 0.158%

references to physical attacks or conflicts

references to mugging incidents and associated discussions

No Comments

No Known Activations