INDEX

Explanations

phrases indicating the implication or consequence of a statement

oai_token-act-pair · gpt-3.5-turbo

instances of the word "meaning" and its variations, indicating explanations or definitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ortex

-0.68

uded

-0.68

ascus

-0.64

ilan

-0.63

ude

-0.62

uminati

-0.61

uding

-0.60

incible

-0.60

 Himself

-0.58

/**

-0.58

POSITIVE LOGITS

 whoever

0.80

 fewer

0.79

 that

0.78

 they

0.77

 goodbye

0.77

 there

0.74

 anyone

0.70

 nobody

0.69

we

0.69

 anything

0.68

Activations Density 0.049%

phrases indicating the implication or consequence of a statement

instances of the word "meaning" and its variations, indicating explanations or definitions

No Comments

No Known Activations