INDEX

Explanations

references to the abbreviation "Am" commonly found in various contexts

oai_token-act-pair · gpt-3.5-turbo

the word "Am" 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 @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ãģį

-0.91

culosis

-0.76

 Titanic

-0.76

theless

-0.72

vous

-0.72

 corners

-0.71

EStreamFrame

-0.66

etheless

-0.66

 learners

-0.65

 Rebels

-0.64

POSITIVE LOGITS

ethyst

1.09

sterdam

1.07

bitious

1.06

nesia

1.02

itan

1.00

utation

0.95

ulet

0.95

essage

0.94

ihara

0.94

ilial

0.92

Activations Density 0.008%

references to the abbreviation "Am" commonly found in various contexts

the word "Am" in various contexts

No Comments

No Known Activations