INDEX

Explanations

words related to leadership positions or control

oai_token-act-pair · gpt-3.5-turbo

terms related to leadership and control positions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

Interstitial

-0.91

Decre

-0.79

 IMAGES

-0.75

Pse

-0.72

Americans

-0.70

ramid

-0.68

Memory

-0.67

birth

-0.66

adra

-0.65

Ident

-0.65

POSITIVE LOGITS

 helm

1.41

idon

0.80

 reins

0.78

fish

0.76

 atop

0.70

 spoiler

0.70

 chair

0.70

eering

0.70

 rigging

0.69

oint

0.69

Activations Density 0.007%

words related to leadership positions or control

terms related to leadership and control positions

No Comments

No Known Activations