INDEX

Explanations

words related to a certain brand starting with "pire" followed by a varying number of characters

oai_token-act-pair · gpt-3.5-turbo

the word "fire" in various forms or contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

iary

-0.78

onent

-0.77

sylv

-0.75

eston

-0.74

¬¼

-0.73

ĨĴ

-0.71

ĸļ

-0.70

arella

-0.67

Ĥ¬

-0.67

acca

-0.66

POSITIVE LOGITS

nces

1.39

lli

1.16

ttes

1.02

lla

1.01

MENTS

0.90

cki

0.85

lessly

0.84

lington

0.83

nder

0.83

tta

0.83

Activations Density 0.071%

words related to a certain brand starting with "pire" followed by a varying number of characters

the word "fire" in various forms or contexts

No Comments

No Known Activations