INDEX

Explanations

phrases indicating someone joining a group or organization

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "joins" and its variations, indicating participation or addition to groups or teams

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

ester

-0.70

framework

-0.69

 tiss

-0.69

mercial

-0.69

 tuber

-0.69

 Merry

-0.68

 appre

-0.66

intendent

-0.62

andi

-0.62

binary

-0.60

POSITIVE LOGITS

joining

0.77

lehem

0.74

 join

0.73

ãĥ³

0.73

iton

0.70

ãĥĥ

0.69

ATURES

0.68

 Join

0.68

 joined

0.66

UCS

0.66

Activations Density 0.022%

phrases indicating someone joining a group or organization

occurrences of the word "joins" and its variations, indicating participation or addition to groups or teams

No Comments

No Known Activations