INDEX

Explanations

phrases related to issuing challenges and confrontations

oai_token-act-pair · gpt-3.5-turbo

words and phrases related to challenging or contesting issues or 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

abet

-0.80

ophe

-0.72

oiler

-0.71

ahon

-0.71

ãĥ¼ãĥĨãĤ£

-0.69

otide

-0.65

fare

-0.64

Goo

-0.63

rie

-0.63

seed

-0.62

POSITIVE LOGITS

enged

0.87

 incumb

0.82

 challengers

0.75

ingly

0.69

 insur

0.68

 authenticity

0.67

 questioning

0.66

 opponents

0.66

ibly

0.66

ioned

0.65

Activations Density 0.048%

phrases related to issuing challenges and confrontations

words and phrases related to challenging or contesting issues or concepts

No Comments

No Known Activations