INDEX

Explanations

text related to news articles and content, possibly focusing on specific sections or paragraphs

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

 sumpay

-0.77

 Himo

-0.68

quarelle

-0.68

 revisor

-0.67

.。

-0.67

 Unito

-0.67

 Allociné

-0.66

 Ause

-0.65

GraphicsUnit

-0.65

 porcentual

-0.65

POSITIVE LOGITS

 shenan

0.71

 McLaugh

0.71

 ineffec

0.70

lxt

0.69

 Considerable

0.69

 philanth

0.68

 impra

0.68

 practition

0.68

 resear

0.67

 Standardi

0.67

Activations Density 10.669%

text related to news articles and content, possibly focusing on specific sections or paragraphs

No Comments

No Known Activations