INDEX

Explanations

references to shelves and shelving in the context of organization or placement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 myſelf

-1.07

Efq

-1.06

 raiſ

-1.02

 Monfieur

-0.99

 purpoſe

-0.99

 pleaſure

-0.96

 Majefty

-0.91

 faſt

-0.90

 seyn

-0.89

 againſt

-0.89

POSITIVE LOGITS

 shelf

0.92

 shelves

0.82

shelf

0.69

item

0.69

0.65

hline

0.61

 Shelf

0.61

 often

0.58

 Shelves

0.55

 Extra

0.55

Activations Density 0.130%

references to shelves and shelving in the context of organization or placement

No Comments

No Known Activations