INDEX

Explanations

mentions or references to the name "iller"

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "Tiller."

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

estinal

-0.75

ðĿ

-0.74

ually

-0.72

aligned

-0.70

elled

-0.69

Gi

-0.67

 Blasio

-0.67

rab

-0.64

¥µ

-0.63

eled

-0.63

POSITIVE LOGITS

espie

1.01

uminati

0.87

aceous

0.86

getic

0.81

culosis

0.77

essee

0.75

iffe

0.74

xit

0.73

adeon

0.73

ussen

0.72

Activations Density 0.018%

mentions or references to the name "iller"

mentions of the word "Tiller."

No Comments

No Known Activations