INDEX

Explanations

instances of the abbreviation "AT&T"

oai_token-act-pair · gpt-3.5-turbo

references to the brand AT&T

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

perse

-0.93

released

-0.77

uyomi

-0.77

ãĥ£

-0.71

ãĥķ

-0.68

clean

-0.67

ãĤ³

-0.67

ovych

-0.66

reviewed

-0.66

 fixme

-0.66

POSITIVE LOGITS

tiny

0.95

AT

0.92

ARI

0.89

GM

0.83

anas

0.81

TAIN

0.81

 least

0.81

mega

0.80

TR

0.77

ributes

0.77

Activations Density 0.008%

instances of the abbreviation "AT&T"

references to the brand AT&T

No Comments

No Known Activations