INDEX

Explanations

significant changes or alterations in direction

oai_token-act-pair · gpt-3.5-turbo

mentions of "course" indicating a direction or path

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

mented

-0.81

rafted

-0.71

alty

-0.66

thening

-0.65

joice

-0.64

bing

-0.64

 stake

-0.63

 curtain

-0.63

ilaterally

-0.62

ascus

-0.61

POSITIVE LOGITS

 Course

1.03

course

1.01

 course

0.99

Course

0.94

 COUR

0.82

hunter

0.77

 courses

0.75

é¾įåĸļå£«

0.71

 Curve

0.69

ibur

0.67

Activations Density 0.013%

significant changes or alterations in direction

mentions of "course" indicating a direction or path

No Comments

No Known Activations