INDEX

Explanations

instances where an action or intention has been consistent or prevalent throughout a situation

oai_token-act-pair · gpt-3.5-turbo

phrases indicating continuity or persistence in actions or experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ahime

-0.76

agra

-0.71

ells

-0.69

ella

-0.63

afer

-0.62

alg

-0.60

ello

-0.60

Redditor

-0.60

=-=-=-=-=-=-=-=-

-0.59

atre

-0.59

POSITIVE LOGITS

MENT

0.81

 thereafter

0.77

ÏĤ

0.72

 except

0.71

 recogn

0.70

 sucked

0.68

 invested

0.68

 succeeded

0.68

Downloadha

0.67

ï¸ı

0.66

Activations Density 0.030%

instances where an action or intention has been consistent or prevalent throughout a situation

phrases indicating continuity or persistence in actions or experiences

No Comments

No Known Activations