INDEX

Explanations

phrases that indicate quantities or counts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

-0.74

Przypisy

-0.69

 purpoſe

-0.69

 Houſe

-0.67

 ſta

-0.66

GEBURTSDATUM

-0.66

verhältnisse

-0.65

 fusca

-0.64

icot

-0.64

 ashamed

-0.63

POSITIVE LOGITS

 ряд

1.19

 several

1.12

 ряда

1.12

several

1.03

 various

0.91

 numerous

0.90

dozen

0.88

Several

0.87

 sejumlah

0.87

いくつか

0.86

Activations Density 0.166%

phrases that indicate quantities or counts

No Comments

No Known Activations