INDEX

Explanations

places or scenes described with detailed narrative elements

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 kram

-1.08

hek

-1.07

 krab

-1.05

 kado

-0.94

 vort

-0.93

 plak

-0.90

 kela

-0.89

 gmbh

-0.88

 blumen

-0.87

 makro

-0.86

POSITIVE LOGITS

 nearby

0.77

 adjacent

0.64

 behind

0.62

 near

0.62

 opposite

0.59

 standing

0.55

nearby

0.52

 beside

0.52

 outside

0.50

 adjoining

0.50

Activations Density 0.327%

places or scenes described with detailed narrative elements

No Comments

No Known Activations