INDEX

Explanations

references to specific rows in a data structure or dataset

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 fubject

-0.87

 poffible

-0.86

 Monfieur

-0.86

 pleaſure

-0.84

 myſelf

-0.82

 occaf

-0.82

 Theſe

-0.82

 ſtill

-0.76

 purpoſe

-0.76

 themſelves

-0.76

POSITIVE LOGITS

row

3.92

Row

3.39

 rows

3.08

Row

3.05

row

3.01

ROW

2.68

 Rows

2.51

ROW

2.36

rows

2.27

Rows

2.17

Activations Density 0.030%

references to specific rows in a data structure or dataset

No Comments

No Known Activations