INDEX

Explanations

commands or instructions preceded by a colon

oai_token-act-pair · gpt-3.5-turbo

colons used to introduce lists or examples

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

odox

-0.76

ritten

-0.72

cially

-0.68

iple

-0.64

orate

-0.63

ificantly

-0.62

sofar

-0.60

¶æ

-0.60

¬

-0.60

rame

-0.59

POSITIVE LOGITS

↵

1.06

<|endoftext|>

1.02

 Firstly

1.01

 âĹı

0.99

↵↵

0.96

↵Âł

0.89

âĹı

0.88

 âĢ¢

0.84

 Â·

0.84

*:

0.83

Activations Density 0.133%

commands or instructions preceded by a colon

colons used to introduce lists or examples

No Comments

No Known Activations