INDEX

Explanations

quotations starting with "That" or "That's"

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

 inext

-1.02

 outlander

-1.01

 dispen

-1.01

 compen

-1.01

 ardu

-1.01

 unce

-1.00

 unil

-1.00

 lein

-0.99

 laun

-0.98

 deleter

-0.98

POSITIVE LOGITS

That

0.80

 That

0.79

 that

0.72

that

0.67

 THAT

0.67

 shouldn

0.66

THAT

0.65

 means

0.61

 wouldn

0.60

 could

0.59

Activations Density 0.170%

quotations starting with "That" or "That's"

No Comments

No Known Activations