INDEX

Explanations

references to a term involving family relations such as "Uncle."

oai_token-act-pair · gpt-3.5-turbo

references to "Uncle" 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 @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

icum

-0.84

istas

-0.81

mberg

-0.72

*/(

-0.70

ties

-0.69

emic

-0.66

ista

-0.65

 Attribution

-0.65

ership

-0.64

izes

-0.63

POSITIVE LOGITS

veland

1.02

 Uncle

0.99

 uncle

0.85

heses

0.83

aned

0.81

 Daddy

0.81

rex

0.79

 princip

0.78

hetic

0.76

arest

0.75

Activations Density 0.014%

references to a term involving family relations such as "Uncle."

references to "Uncle" in various contexts

No Comments

No Known Activations