INDEX

Explanations

abbreviations, symbols, and characters arranged in a specific pattern

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 betweenstory

-0.84

SourceChecksum

-0.70

localctx

-0.67

 <>",

-0.63

awtextra

-0.63

TestTools

-0.60

stateProvider

-0.60

parsedMessage

-0.60

WriteBarrier

-0.58

 SEDS

-0.58

POSITIVE LOGITS

 unlaw

0.98

 pamph

0.96

 encomp

0.90

 indestru

0.86

 forbear

0.79

 reconno

0.78

 shenan

0.78

 Whence

0.78

 viciss

0.77

 scrat

0.77

Activations Density 0.087%

abbreviations, symbols, and characters arranged in a specific pattern

No Comments

No Known Activations