INDEX

Explanations

names of specific individuals

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals or organizations

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

ysis

-0.82

agy

-0.77

er

-0.76

usc

-0.75

safe

-0.73

ership

-0.72

hang

-0.72

sticks

-0.70

itiz

-0.70

ocity

-0.70

POSITIVE LOGITS

 Fraser

0.88

incinn

0.77

forth

0.75

 Opposition

0.72

atoon

0.72

regor

0.69

EStream

0.68

PLA

0.67

 clan

0.67

waters

0.66

Activations Density 0.032%

names of specific individuals

references to specific individuals or organizations

No Comments

No Known Activations