INDEX

Explanations

mentions of a specific name, "Warren", likely referring to the name of a person or character in the text

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

PerformLayout

-0.50

 Kirkwood

-0.50

 esternos

-0.50

IANGLES

-0.49

 povezave

-0.48

 Medford

-0.48

 Lockhart

-0.48

 Malden

-0.48

colat

-0.47

Jeśli

-0.46

POSITIVE LOGITS

 Warren

1.45

Warren

1.40

 WARREN

1.32

WARR

1.04

 prek

0.87

 franz

0.82

 loren

0.78

 Traité

0.78

 Gouver

0.77

 leonardo

0.75

Activations Density 0.149%

mentions of a specific name, "Warren", likely referring to the name of a person or character in the text

No Comments

No Known Activations