INDEX

Explanations

phrases related to reasoning or logic, especially regarding biases and fallacies

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

 inev

-1.22

 accla

-1.18

 encomp

-1.17

 affor

-1.13

 increa

-1.12

 reluct

-1.11

 volunte

-1.10

 maneu

-1.08

 intersper

-1.07

 excru

-1.06

POSITIVE LOGITS

 nothing

0.96

nothing

0.88

Nothing

0.82

 Nothing

0.79

 rather

0.77

 never

0.74

no

0.73

not

0.73

 ignore

0.72

 nowhere

0.71

Activations Density 0.558%

phrases related to reasoning or logic, especially regarding biases and fallacies

No Comments

No Known Activations