INDEX

Explanations

phrases or sentences indicating a significant event or milestone

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "marked" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ADS

-0.73

ettel

-0.70

isance

-0.70

adra

-0.69

amia

-0.67

urus

-0.66

 listened

-0.66

asion

-0.65

 ILCS

-0.65

rouch

-0.65

POSITIVE LOGITS

 marked

0.87

manship

0.84

 marking

0.81

 marks

0.81

ocument

0.78

mast

0.78

 signs

0.75

paren

0.72

marked

0.72

 markings

0.72

Activations Density 0.012%

phrases or sentences indicating a significant event or milestone

occurrences of the word "marked" in various contexts

No Comments

No Known Activations