INDEX

Explanations

mentions of the word "ola."

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Lola"

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

ergy

-0.75

wcsstore

-0.74

frames

-0.71

picking

-0.67

ERAL

-0.66

rish

-0.66

sworth

-0.66

rings

-0.65

rant

-0.65

frame

-0.64

POSITIVE LOGITS

onga

0.94

iffe

0.93

Ã±o

0.91

BILITY

0.91

isi

0.89

ppe

0.89

zzle

0.87

zzi

0.87

uthor

0.86

uga

0.83

Activations Density 0.010%

mentions of the word "ola."

instances of the word "Lola"

No Comments

No Known Activations