INDEX

Explanations

the word "that" in sentences, possibly indicating a focus on specific contexts or conditions

oai_token-act-pair · gpt-3.5-turbo

the word "that" to identify clauses or statements

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

Guard

-0.67

oses

-0.66

lean

-0.65

gur

-0.64

respect

-0.63

aq

-0.63

le

-0.60

ounding

-0.59

´

-0.58

van

-0.58

POSITIVE LOGITS

 they

0.92

soever

0.89

 THEY

0.84

 there

0.83

we

0.81

 unlike

0.79

*/(

0.78

 although

0.78

 nobody

0.78

it

0.70

Activations Density 0.116%

the word "that" in sentences, possibly indicating a focus on specific contexts or conditions

the word "that" to identify clauses or statements

No Comments

No Known Activations