INDEX

Explanations

phrases related to evaluation, judgment, or categorization

oai_token-act-pair · gpt-3.5-turbo

instances of the word "deem" and its variations, indicating assessments or judgments

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

 Sparrow

-0.70

 Rapp

-0.66

 scrimmage

-0.65

 Pacers

-0.63

dayName

-0.62

 Hawks

-0.61

Orn

-0.61

irgin

-0.60

 Lawn

-0.59

 Healthcare

-0.59

POSITIVE LOGITS

phas

1.41

EMP

0.85

 deem

0.82

 unemploy

0.77

leted

0.75

 unfit

0.75

 discretion

0.73

eme

0.73

lessly

0.72

rul

0.71

Activations Density 0.035%

phrases related to evaluation, judgment, or categorization

instances of the word "deem" and its variations, indicating assessments or judgments

No Comments

No Known Activations