INDEX

Explanations

statements about beliefs and convictions

oai_token-act-pair · gpt-3.5-turbo

expressions related to belief

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

soever

-0.82

killed

-0.70

Ware

-0.62

 wipes

-0.62

CVE

-0.62

MAP

-0.61

hai

-0.61

 preferably

-0.61

 hooks

-0.61

 aptly

-0.60

POSITIVE LOGITS

clus

1.28

clusions

1.09

animate

0.91

clusively

0.90

organic

0.85

 accordance

0.85

effic

0.84

 favor

0.84

 unison

0.83

abilities

0.83

Activations Density 0.095%

statements about beliefs and convictions

expressions related to belief

No Comments

No Known Activations