INDEX

Explanations

phrases related to investigations or probes

oai_token-act-pair · gpt-3.5-turbo

references to investigations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

birth

-0.74

 Distribut

-0.73

Unix

-0.66

emonic

-0.66

ãĤª

-0.65

yip

-0.64

arus

-0.62

 penn

-0.61

veland

-0.61

ingers

-0.61

POSITIVE LOGITS

 into

1.31

 INTO

1.13

into

1.09

 probing

1.04

 probe

1.03

 Into

1.00

 probes

0.96

 undertaken

0.96

 conducted

0.95

 investigation

0.95

Activations Density 0.080%

phrases related to investigations or probes

references to investigations

No Comments

No Known Activations