INDEX

Explanations

information related to military ranks and accomplishments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

utherland

-0.17

ãĥ¯ãĥ¼

-0.15

overnment

-0.15

atron

-0.15

igi

-0.14

 circular

-0.14

arda

-0.13

 Bucket

-0.13

 Îĸ

-0.13

aller

-0.13

POSITIVE LOGITS

 rank

0.28

rank

0.23

 ranks

0.22

 Rank

0.21

 promot

0.20

 senior

0.20

 promotion

0.20

.rank

0.19

 Senior

0.18

 promoted

0.18

Activations Density 0.061%

information related to military ranks and accomplishments

No Comments

No Known Activations