INDEX

Explanations

long-term actions or experiences that individuals have been involved in

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "I’ve been."

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

izable

-0.71

Profile

-0.69

lies

-0.68

ierrez

-0.66

ulence

-0.66

 Tycoon

-0.66

rouse

-0.64

iop

-0.64

warts

-0.64

Must

-0.64

POSITIVE LOGITS

 able

1.05

 bitten

0.99

 subjected

0.99

 unable

0.98

 briefed

0.93

cffffcc

0.91

 forgiven

0.90

 fortunate

0.90

 notified

0.89

 contacted

0.89

Activations Density 0.130%

long-term actions or experiences that individuals have been involved in

instances of the phrase "I’ve been."

No Comments

No Known Activations