INDEX

Explanations

references to government ministries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

abis

-0.15

odd

-0.14

orca

-0.14

irt

-0.14

ÑħÐ°

-0.14

 Wend

-0.13

atica

-0.13

rega

-0.13

 subgroup

-0.13

POSITIVE LOGITS

ombat

0.18

ider

0.16

oggled

0.16

uge

0.15

dens

0.14

fts

0.14

legg

0.13

rans

0.13

 blasting

0.13

 finance

0.13

Activations Density 0.032%

references to government ministries

No Known Activations