INDEX

Explanations

references to flags

oai_token-act-pair · gpt-3.5-turbo

references to flags and their significance, particularly in relation to patriotism

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

 Argon

-0.72

 Flavoring

-0.71

è£ħ

-0.71

nesota

-0.70

axter

-0.68

izabeth

-0.67

 subsequ

-0.66

venants

-0.66

Kut

-0.65

 advers

-0.63

POSITIVE LOGITS

pole

1.53

rant

1.45

ging

1.42

staff

1.25

 bearer

1.06

ged

1.01

gers

0.99

stones

0.98

rants

0.98

rier

0.93

Activations Density 0.036%

references to flags

references to flags and their significance, particularly in relation to patriotism

No Comments

No Known Activations