INDEX

Explanations

words related to geographical locations, specifically "Liv" and "Lov" appearing frequently in texts

oai_token-act-pair · gpt-3.5-turbo

terms related to living and love

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 shroud

-0.70

 reluct

-0.66

porting

-0.65

ported

-0.64

uyomi

-0.62

 contraceptives

-0.61

 cryst

-0.61

 restraints

-0.60

 buggy

-0.60

*/(

-0.60

POSITIVE LOGITS

iew

0.89

ocity

0.84

ership

0.83

ousy

0.82

est

0.82

oret

0.81

ocate

0.81

ocations

0.80

actory

0.79

igation

0.78

Activations Density 0.023%

words related to geographical locations, specifically "Liv" and "Lov" appearing frequently in texts

terms related to living and love

No Comments

No Known Activations