INDEX

Explanations

time-related phrases and durations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

UnitTesting

-0.73

mappedBy

-0.70

 springfox

-0.70

slidesPer

-0.68

 autorytatywna

-0.67

ErrorListener

-0.66

DebuggerNonUser

-0.63

 BoxDecoration

-0.62

PhysRev

-0.60

ActionCreators

-0.59

POSITIVE LOGITS

 overnight

1.24

 Overnight

1.06

 weekend

1.02

Overnight

1.00

vernight

0.96

 morning

0.92

 yesterday

0.87

yesterday

0.86

weekend

0.85

 tomorrow

0.84

Activations Density 0.262%

time-related phrases and durations

No Comments

No Known Activations