INDEX

Explanations

the word "Ye" in different contexts

oai_token-act-pair · gpt-3.5-turbo

references to the entity "Ye," indicating a focus on mentions of a specific person or character

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĥ¼ãĥĨ

-1.01

Downloadha

-0.81

ITY

-0.76

ctica

-0.75

ãĥ´ãĤ¡

-0.73

ãĥ¼ãĥĨãĤ£

-0.73

lished

-0.71

ity

-0.69

ãĥĨ

-0.68

Ø§ÙĦ

-0.68

POSITIVE LOGITS

ldon

1.01

ats

0.92

lda

0.89

oman

0.88

erk

0.84

aser

0.81

asing

0.81

igh

0.80

asers

0.79

hovah

0.77

Activations Density 0.012%

the word "Ye" in different contexts

references to the entity "Ye," indicating a focus on mentions of a specific person or character

No Comments

No Known Activations