INDEX

Explanations

relationships or conflicts between different entities or groups

oai_token-act-pair · gpt-3.5-turbo

references to interpersonal or intergroup conflicts and tensions

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

Coy

-0.73

jew

-0.71

Downloadha

-0.65

ï¸

-0.64

anwhile

-0.64

tre

-0.63

 Didn

-0.62

minist

-0.61

hiro

-0.61

 meanwhile

-0.59

POSITIVE LOGITS

 alike

0.69

 extremes

0.67

 throats

0.65

 versa

0.62

ãĤ¡

0.61

 viewpoints

0.59

ãĤ©

0.58

 circulation

0.58

 respective

0.58

 assailants

0.57

Activations Density 0.306%

relationships or conflicts between different entities or groups

references to interpersonal or intergroup conflicts and tensions

No Comments

No Known Activations