INDEX

Explanations

words related to professions and professionals

oai_token-act-pair · gpt-3.5-turbo

references to various types of professionals in different fields

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

 mileage

-0.65

 Territories

-0.64

 streaks

-0.62

apo

-0.62

Loop

-0.62

seed

-0.61

 Plate

-0.58

 indefinite

-0.57

Pause

-0.56

Phones

-0.56

POSITIVE LOGITS

paces

1.01

umers

0.95

ervative

0.93

hips

0.90

ervatives

0.90

 specialize

0.86

 specializing

0.84

who

0.84

bourg

0.81

folk

0.79

Activations Density 0.049%

words related to professions and professionals

references to various types of professionals in different fields

No Comments

No Known Activations