INDEX

Explanations

proper nouns and titles in a corporate context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

 assistants

-0.19

usch

-0.15

 assistant

-0.15

ensis

-0.15

ÏĮÏĦÎ·ÏĦÎ±ÏĤ

-0.14

assistant

-0.14

èĩ£

-0.14

 orta

-0.14

Assistant

-0.14

conc

-0.14

POSITIVE LOGITS

 president

0.35

CEO

0.34

 President

0.33

CEO

0.31

President

0.28

 chairman

0.26

 presidente

0.25

 prÃ©sident

0.25

-president

0.25

 Chairman

0.24

Activations Density 0.095%

proper nouns and titles in a corporate context

No Comments

No Known Activations