INDEX

Explanations

words related to academic disciplines or fields of study, particularly focused on "Sciences."

oai_token-act-pair · gpt-3.5-turbo

references to academic programs or departments related to sciences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

usher

-0.71

shr

-0.65

ints

-0.64

spir

-0.64

rel

-0.64

gran

-0.63

ieft

-0.62

own

-0.61

adic

-0.61

gotten

-0.60

POSITIVE LOGITS

 Sciences

1.14

ource

0.94

 sciences

0.83

 Neuroscience

0.82

 Techn

0.79

 Instrument

0.79

creen

0.78

 Journalism

0.76

 Applications

0.76

kies

0.75

Activations Density 0.014%

words related to academic disciplines or fields of study, particularly focused on "Sciences."

references to academic programs or departments related to sciences

No Comments

No Known Activations