INDEX

Explanations

terms related to modes and references to planets

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Efq

-1.80

 itſelf

-1.69

 myſelf

-1.67

 Monfieur

-1.62

 Jefus

-1.61

 pleaſure

-1.55

 Theſe

-1.54

 Reſ

-1.51

 ſeveral

-1.47

 themſelves

-1.45

POSITIVE LOGITS

mode

1.27

MODE

1.07

 mode

1.02

<eos>

0.95

0.91

↵↵

0.85

0.83

Mode

0.82

ש

0.82

↵

0.81

Activations Density 0.129%

terms related to modes and references to planets

No Comments

No Known Activations