INDEX

Explanations

text fragments expressing emotions, reactions, or personal evaluations

oai_token-act-pair · gpt-3.5-turbo

expressions of personal emotions and feelings

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

aceutical

-0.64

STA

-0.61

othing

-0.61

 Pigs

-0.61

 bombshell

-0.60

rising

-0.59

 TEXT

-0.58

Times

-0.58

ufact

-0.57

earchers

-0.57

POSITIVE LOGITS

 ineligible

0.84

 feel

0.77

 uneasy

0.73

 eligible

0.73

 appealing

0.72

 susceptible

0.72

 safer

0.70

 seem

0.69

 hesitate

0.69

 reconsider

0.69

Activations Density 0.086%

text fragments expressing emotions, reactions, or personal evaluations

expressions of personal emotions and feelings

No Comments

No Known Activations