INDEX

Explanations

references to chains or sequences, particularly in a legal or procedural context

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

 myſelf

-0.97

 juſt

-0.89

 deſt

-0.88

 Cabrio

-0.87

 themſelves

-0.86

uſ

-0.86

 Monfieur

-0.85

 auffi

-0.84

 uſed

-0.84

 obſ

-0.83

POSITIVE LOGITS

 chain

2.88

 chains

2.65

chain

2.49

 Chain

2.46

 CHAIN

2.38

Chain

2.32

 Chains

2.21

chains

2.19

CHAIN

2.05

Chains

1.99

Activations Density 0.038%

references to chains or sequences, particularly in a legal or procedural context

No Comments

No Known Activations