INDEX

Explanations

phrases instructing or prompting the reader to check back for updates or additional information

oai_token-act-pair · gpt-3.5-turbo

references to checking emails and updates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

EStreamFrame

-0.71

rawdownloadcloneembedreportprint

-0.65

von

-0.62

å§«

-0.62

 Learns

-0.61

©¶æ¥µ

-0.61

ahime

-0.61

ĸļ

-0.60

ĨĴ

-0.58

 Bucc

-0.58

POSITIVE LOGITS

 balances

0.76

 sidelines

0.73

bones

0.73

worthiness

0.72

geist

0.70

balance

0.70

 periphery

0.67

 compatibility

0.67

mate

0.64

security

0.63

Activations Density 0.101%

phrases instructing or prompting the reader to check back for updates or additional information

references to checking emails and updates

No Comments

No Known Activations