INDEX

Explanations

words related to concepts or things that are beyond the usual, expected, or normal

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "beyond."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

yah

-0.70

 Assass

-0.70

pmwiki

-0.67

è¦ļéĨĴ

-0.66

baugh

-0.66

rar

-0.65

RAW

-0.64

ARS

-0.63

cu

-0.62

ICAN

-0.62

POSITIVE LOGITS

 comprehension

0.77

 Greece

0.75

 beyond

0.73

esis

0.73

ordinary

0.72

lehem

0.70

ĸļ

0.69

dfx

0.69

=>

0.68

atos

0.67

Activations Density 0.015%

words related to concepts or things that are beyond the usual, expected, or normal

occurrences of the word "beyond."

No Comments

No Known Activations