INDEX

Explanations

phrases related to social and political issues, particularly focused on perspectives and opinions

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

 potest

-0.78

 makro

-0.75

 habet

-0.74

 balon

-0.72

<bos>

-0.70

 pels

-0.69

 poliuret

-0.67

 polig

-0.65

 veda

-0.65

 lampa

-0.64

POSITIVE LOGITS

 unspeak

1.23

 apprehen

1.08

 shenan

1.04

 indescri

1.02

 gild

1.00

 intersper

1.00

 gaily

1.00

 reluct

1.00

 perchance

0.99

 snoopy

0.96

Activations Density 0.243%

phrases related to social and political issues, particularly focused on perspectives and opinions

No Comments

No Known Activations