INDEX

Explanations

terms related to confessions or admitting guilt

oai_token-act-pair · gpt-3.5-turbo

terms related to admissions or declarations of accountability

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

 Cheong

-0.67

::::::::

-0.64

 Orchestra

-0.62

 upkeep

-0.62

axy

-0.62

 brackets

-0.61

undai

-0.60

 braces

-0.60

chnology

-0.60

roups

-0.59

POSITIVE LOGITS

 confess

1.01

 confessions

0.94

itives

0.86

 confession

0.84

 confessed

0.82

essors

0.75

ification

0.73

essional

0.73

ities

0.72

itive

0.70

Activations Density 0.017%

terms related to confessions or admitting guilt

terms related to admissions or declarations of accountability

No Comments

No Known Activations