INDEX

Explanations

references to marine animals, particularly those related to whale watching and interactions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 Damascus

-0.59

 EconPapers

-0.54

ymce

-0.53

modium

-0.53

ochial

-0.53

 ferns

-0.52

ICULO

-0.52

pruch

-0.52

cektir

-0.51

dirt

-0.51

POSITIVE LOGITS

 whale

1.09

 whales

1.06

 Whale

0.99

Whale

0.92

 Whales

0.91

whale

0.87

 marine

0.86

 dolphins

0.84

 ocean

0.80

🐋

0.77

Activations Density 0.148%

references to marine animals, particularly those related to whale watching and interactions

No Comments

No Known Activations