INDEX

Explanations

words expressing comparisons and newness

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

 Augu

-0.86

 bangkok

-0.81

 stockholm

-0.79

 venice

-0.77

 Gorb

-0.77

 Riti

-0.76

 lidl

-0.76

fta

-0.75

 linden

-0.74

 Idem

-0.74

POSITIVE LOGITS

 rarely

0.67

 never

0.65

 hadn

0.65

 seldom

0.62

 lately

0.61

 recently

0.56

 until

0.55

never

0.53

 hasn

0.52

 haven

0.51

Activations Density 0.232%

words expressing comparisons and newness

No Comments

No Known Activations