INDEX

Explanations

specific proper nouns and their associated attributes in a given context

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ness

-1.02

nya

-0.84

으로

-0.67

night

-0.65

-0.64

real

-0.61

scan

-0.61

road

-0.59

research

-0.58

rings

-0.58

POSITIVE LOGITS

0.76

eeee

0.70

eeeeee

0.64

eee

0.64

eeeeeeee

0.63

xtra

0.63

EEEE

0.61

EEEEEE

0.60

eeeee

0.60

 asyncio

0.59

Activations Density 0.434%

specific proper nouns and their associated attributes in a given context

No Comments

No Known Activations