INDEX

Explanations

terms related to disagreement or negation in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 AssemblyProduct

-0.73

tà

-0.73

SuppressLint

-0.73

orianCalendar

-0.69

 pleins

-0.67

Varint

-0.67

ويد

-0.67

omitempty

-0.65

Wey

-0.64

bindValue

-0.64

POSITIVE LOGITS

Dis

2.20

Dis

2.17

dis

2.13

DIS

2.08

DIS

1.95

 dispersion

1.55

dis

1.50

 dispens

1.47

 Dispersion

1.46

 disp

1.41

Activations Density 0.160%

terms related to disagreement or negation in various contexts

No Comments

No Known Activations