INDEX

Explanations

the word "mere" followed by a number

oai_token-act-pair · gpt-3.5-turbo

the word "mere" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 smartest

-0.65

illas

-0.64

 Also

-0.61

 Cursed

-0.60

 Locked

-0.60

multipl

-0.58

ibal

-0.58

lined

-0.58

 whichever

-0.57

 Investors

-0.55

POSITIVE LOGITS

 mortals

1.19

 incidental

0.86

 superficial

0.84

ifiable

0.75

 consolation

0.74

 curs

0.74

 curiosity

0.73

 inconven

0.73

 cosmetic

0.73

 coincidence

0.72

Activations Density 0.032%

the word "mere" followed by a number

the word "mere" in various contexts

No Comments

No Known Activations