INDEX

Explanations

terms related to errors or misclassifications

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

InputBorder

-0.92

InjectMocks

-0.90

 חיצוניים

-0.85

 LCCN

-0.83

horabuena

-0.83

 tramonto

-0.80

ьаж

-0.79

AccessorTable

-0.78

WireFormatLite

-0.77

FieldNumber

-0.77

POSITIVE LOGITS

mis

2.34

Mis

2.12

Mis

2.11

mis

2.07

MIS

2.00

MIS

1.72

 mismatch

1.24

 misuse

1.23

 misused

1.18

 misc

1.16

Activations Density 0.050%

terms related to errors or misclassifications

No Comments

No Known Activations