INDEX

Explanations

specific items in a list

oai_token-act-pair · gpt-3.5-turbo

colons followed by lists or items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

gow

-0.65

orate

-0.64

oland

-0.63

ashington

-0.62

pty

-0.61

veland

-0.59

agon

-0.58

iliate

-0.58

ritten

-0.58

hed

-0.57

POSITIVE LOGITS

<|endoftext|>

1.15

↵

1.14

 âĹı

1.08

 âĢ¢

1.04

↵↵

1.03

 Â·

0.99

 Firstly

0.93

↵Âł

0.90

âĢ¢

0.87

âĹı

0.86

Activations Density 0.120%

specific items in a list

colons followed by lists or items

No Comments

No Known Activations