INDEX

Explanations

programming-related keywords and syntax

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

ÑħÑĥ

-0.17

stral

-0.17

usercontent

-0.16

Dll

-0.16

 çĻ¾åº¦

-0.15

exam

-0.14

acre

-0.14

.Double

-0.14

 Demon

-0.14

udad

-0.13

POSITIVE LOGITS

 Bench

0.17

ouch

0.15

Lon

0.14

ARAM

0.14

mam

0.14

 Vers

0.14

 grenade

0.14

 chung

0.14

0.13

 issued

0.13

Activations Density 0.002%

programming-related keywords and syntax

No Comments

No Known Activations