INDEX

Explanations

punctuation marks indicating a question in a sentence

oai_token-act-pair · gpt-3.5-turbo

punctuation marks, specifically colons, indicating questions or significant statements

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

izons

-0.79

ema

-0.76

onent

-0.72

onet

-0.71

uced

-0.70

 carbohyd

-0.70

oreal

-0.69

 tremend

-0.69

ulent

-0.69

alysed

-0.68

POSITIVE LOGITS

Bye

0.86

Why

0.81

 "âĢ¦

0.81

 namely

0.78

 Forget

0.74

Who

0.74

How

0.74

 Does

0.73

WHY

0.72

 What

0.72

Activations Density 0.123%

punctuation marks indicating a question in a sentence

punctuation marks, specifically colons, indicating questions or significant statements

No Comments

No Known Activations