INDEX

Explanations

references to errors and warnings, especially related to file or access issues

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

 houſe

-0.99

 wikipagina

-0.99

 pleaſure

-0.96

 Houſe

-0.93

 ſche

-0.92

 purpoſe

-0.92

 Shakspeare

-0.86

 Majefty

-0.86

 raiſ

-0.84

 Cæsar

-0.83

POSITIVE LOGITS

 same

1.11

 entire

1.05

 most

1.03

 following

0.92

 whole

0.90

 latter

0.89

 rest

0.89

 main

0.86

“

0.84

 latest

0.81

Activations Density 0.007%

references to errors and warnings, especially related to file or access issues

No Comments

No Known Activations