INDEX

Explanations

words related to emotional or psychological states

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

torch.float32

Hook Point

blocks.8.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

manship

-0.61

minecraft

-0.55

choice

-0.54

 offense

-0.54

 Admir

-0.54

 Cruiser

-0.53

ocular

-0.53

addons

-0.53

 Boxing

-0.53

gain

-0.53

POSITIVE LOGITS

cially

0.86

agog

0.79

llah

0.77

0.72

ciating

0.71

oult

0.68

cial

0.68

cific

0.67

Ö¼

0.67

ruct

0.65

Activations Density 0.019%

words related to emotional or psychological states

No Comments

No Known Activations