INDEX

Explanations

phrases that express conditions or consequences

oai_token-act-pair · gpt-3.5-turbo

conditional statements or phrases that introduce hypothetical scenarios

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

enos

-0.68

akia

-0.67

 Pledge

-0.66

hs

-0.64

azo

-0.63

urga

-0.62

FTWARE

-0.59

Gov

-0.59

 Fixes

-0.59

 Sins

-0.59

POSITIVE LOGITS

fy

1.21

 indeed

0.99

 ever

0.97

 anything

0.95

not

0.95

tar

0.89

rame

0.88

 possible

0.83

ever

0.79

 unchecked

0.77

Activations Density 0.064%

phrases that express conditions or consequences

conditional statements or phrases that introduce hypothetical scenarios

No Comments

No Known Activations