INDEX

Explanations

phrases related to specific locations or events

oai_token-act-pair · gpt-3.5-turbo

geographical locations and urban terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 guiActiveUn

-0.74

 Mour

-0.73

vironment

-0.67

©¶æ

-0.67

ovych

-0.63

£ı

-0.63

phies

-0.63

Ĭ±

-0.62

ģĸ

-0.62

nih

-0.62

POSITIVE LOGITS

hett

0.74

aneously

0.69

abis

0.69

olini

0.69

atoon

0.68

essim

0.68

oslav

0.67

lator

0.65

ooter

0.64

PB

0.64

Activations Density 0.066%

phrases related to specific locations or events

geographical locations and urban terminology

No Comments

No Known Activations