INDEX

Explanations

words related to a reversible or opposite concept

oai_token-act-pair · gpt-3.5-turbo

references to various types of "verses" or poetic structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

urated

-0.86

arij

-0.78

RAW

-0.72

raped

-0.68

uracy

-0.68

Kut

-0.65

URA

-0.65

ritz

-0.65

istered

-0.65

icum

-0.64

POSITIVE LOGITS

verse

1.00

lihood

0.93

terday

0.85

theless

0.84

itudinal

0.70

vous

0.69

engers

0.69

vable

0.69

er

0.67

 Collider

0.67

Activations Density 0.015%

words related to a reversible or opposite concept

references to various types of "verses" or poetic structures

No Comments

No Known Activations