INDEX

Explanations

unique identifiers or specific codes related to a series or show

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ÐĲÑĢÑħÑĸÐ²

-0.20

cts

-0.16

hsi

-0.16

Å¥

-0.15

ataka

-0.15

vanished

-0.14

iedo

-0.14

mrt

-0.14

ì¦Ŀê¸Ī

-0.14

chen

-0.14

POSITIVE LOGITS

ãħ

0.15

 ãħ

0.15

ìļĶ

0.15

ÃŃ

0.14

 Ãºn

0.14

jeg

0.14

 unlocked

0.14

¤í

0.14

 rÃ¡

0.13

Ung

0.13

Activations Density 0.001%

unique identifiers or specific codes related to a series or show

No Comments

No Known Activations