INDEX

Explanations

specific patterns in text matching a certain format, possibly related to coding or a specific type of organization

oai_token-act-pair · gpt-3.5-turbo

occurrences of a specific character or symbol

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 scattering

-0.75

 ende

-0.62

 puff

-0.62

 helpless

-0.61

 maximum

-0.61

 shadow

-0.61

 incorpor

-0.60

 unconscious

-0.60

 sprawling

-0.60

 acknow

-0.60

POSITIVE LOGITS

ï¸ı

1.13

Â¯

1.06

0.93

ef

0.85

tm

0.85

efe

0.83

ti

0.83

0.82

sch

0.81

culosis

0.81

Activations Density 0.156%

specific patterns in text matching a certain format, possibly related to coding or a specific type of organization

occurrences of a specific character or symbol

No Comments

No Known Activations