INDEX

Explanations

terms related to conflict and competition

oai_token-act-pair · gpt-3.5-turbo

terms related to competitive or conflict-driven situations

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

Lod

-0.62

Fac

-0.62

 Regist

-0.61

express

-0.61

 slightest

-0.59

disabled

-0.58

AFB

-0.58

ãģį

-0.57

 Reference

-0.57

imm

-0.56

POSITIVE LOGITS

naire

0.87

ousel

0.86

ilogy

0.81

ernaut

0.78

 frenzy

0.76

 revolving

0.74

 spree

0.72

 mentality

0.69

storm

0.68

 spearheaded

0.68

Activations Density 0.322%

terms related to conflict and competition

terms related to competitive or conflict-driven situations

No Comments

No Known Activations