INDEX

Explanations

mentions of professionals or occupations, specifically engineers

oai_token-act-pair · gpt-3.5-turbo

references to various types of engineers

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

vic

-0.79

si

-0.75

aver

-0.75

berra

-0.74

ibaba

-0.72

eworld

-0.71

Interstitial

-0.67

emonic

-0.66

emb

-0.66

ints

-0.65

POSITIVE LOGITS

 engineer

1.08

 technician

0.85

rador

0.83

 Engineer

0.81

urally

0.81

 architect

0.79

woman

0.77

 biologist

0.76

 engineers

0.75

 designer

0.74

Activations Density 0.007%

mentions of professionals or occupations, specifically engineers

references to various types of engineers

No Comments

No Known Activations