INDEX

Explanations

phrases related to breaches, violations, and misconduct

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "breach" and related phrases indicating violations or infractions

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

uana

-0.79

lee

-0.69

 livest

-0.67

 Towns

-0.66

eele

-0.65

atable

-0.63

 redistributed

-0.63

 unification

-0.61

puff

-0.61

 compr

-0.60

POSITIVE LOGITS

 breaches

0.93

aucus

0.85

 breach

0.83

 Breach

0.79

terness

0.78

 breached

0.72

hold

0.70

ionage

0.69

mson

0.68

ware

0.68

Activations Density 0.014%

phrases related to breaches, violations, and misconduct

occurrences of the term "breach" and related phrases indicating violations or infractions

No Comments

No Known Activations