INDEX

Explanations

phrases related to significant actions or decisions

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

entimes

-0.66

erest

-0.63

ãĤ©

-0.62

ibaba

-0.61

 excel

-0.60

igree

-0.60

 adversity

-0.60

estial

-0.59

ensity

-0.57

IBLE

-0.57

POSITIVE LOGITS

 indefinitely

1.52

 altogether

1.37

due

1.36

due

1.29

 pending

1.25

 citing

1.25

until

1.14

 because

1.12

 owing

1.10

 amid

1.09

Activations Density 0.589%

phrases related to significant actions or decisions

No Comments

No Known Activations