INDEX

Explanations

specific instances related to decision making or personal experiences

oai_token-act-pair · gpt-3.5-turbo

phrases indicating decision-making or consideration processes

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

apo

-0.70

ROR

-0.62

 Witches

-0.61

DN

-0.59

photos

-0.56

Lago

-0.56

uggle

-0.55

role

-0.55

ombat

-0.55

 fixme

-0.55

POSITIVE LOGITS

 finally

1.27

 settled

1.01

 decided

0.99

 conclude

0.90

 eventually

0.89

 reluctantly

0.87

 culminating

0.85

 recons

0.83

 concluded

0.83

 decide

0.82

Activations Density 0.360%

specific instances related to decision making or personal experiences

phrases indicating decision-making or consideration processes

No Comments

No Known Activations