INDEX

Explanations

proper nouns or names containing the word "Nob" or "Noble"

oai_token-act-pair · gpt-3.5-turbo

references to specific names and organizations

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

ization

-0.82

 Nadu

-0.77

utic

-0.77

ablishment

-0.74

ize

-0.73

icago

-0.71

 Communication

-0.66

izing

-0.65

ized

-0.65

izable

-0.64

POSITIVE LOGITS

doms

0.91

edience

0.88

odies

0.85

lique

0.85

Nob

0.84

ility

0.84

inished

0.83

leness

0.80

bey

0.78

poon

0.78

Activations Density 0.032%

proper nouns or names containing the word "Nob" or "Noble"

references to specific names and organizations

No Comments

No Known Activations