INDEX

Explanations

phrases related to making statements or declarations

oai_token-act-pair · gpt-3.5-turbo

instances of the word "proclaim" 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 @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

pai

-0.75

lua

-0.73

ORN

-0.73

redd

-0.71

Unix

-0.71

urai

-0.71

Factor

-0.70

Ń·

-0.69

MpServer

-0.68

EPA

-0.67

POSITIVE LOGITS

 Ange

0.80

 Koen

0.71

anu

0.69

 Crusade

0.69

 Sicily

0.69

 Dinosaur

0.68

 Everest

0.67

Leo

0.67

 Gamb

0.66

 Lans

0.66

Activations Density 0.000%

phrases related to making statements or declarations

instances of the word "proclaim" in various contexts

No Comments

No Known Activations