INDEX

Explanations

numbers in sequences, possibly related to measurements or quantities

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "for" and similar prepositions indicating relationships or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

press

-0.76

WER

-0.71

meet

-0.70

DragonMagazine

-0.70

Previously

-0.70

uid

-0.68

WHAT

-0.67

Press

-0.66

ogical

-0.66

POSITIVE LOGITS

TBA

0.79

 Hots

0.70

 Coff

0.69

 Thur

0.67

 Shap

0.67

 Typhoon

0.66

 Mour

0.66

 Gren

0.66

 Sapp

0.65

Ec

0.65

Activations Density 0.291%

numbers in sequences, possibly related to measurements or quantities

occurrences of the word "for" and similar prepositions indicating relationships or conditions

No Comments

No Known Activations