INDEX

Explanations

references to a specific name, possibly "Obama", as well as words related to legislation and policy

oai_token-act-pair · gpt-3.5-turbo

references to "Obamacare" and related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

theless

-0.78

 Remastered

-0.72

WAYS

-0.69

 indoors

-0.68

fare

-0.67

rawdownloadcloneembedreportprint

-0.66

 chorus

-0.65

ãĥ¼ãĥĨãĤ£

-0.64

 à¨

-0.64

âĸ¬

-0.63

POSITIVE LOGITS

taining

1.30

solete

1.21

fusc

1.14

tained

1.05

tain

1.01

noxious

1.01

amacare

0.98

lig

0.94

scene

0.94

amac

0.93

Activations Density 0.033%

references to a specific name, possibly "Obama", as well as words related to legislation and policy

references to "Obamacare" and related concepts

No Comments

No Known Activations