INDEX

Explanations

conjunctions, particularly the word "and"

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ited

-0.16

opus

-0.15

ãĤ¤ãĤº

-0.14

.RunWith

-0.14

ugins

-0.14

ucch

-0.14

 latter

-0.14

velt

-0.14

ital

-0.14

POSITIVE LOGITS

/or

0.17

ehr

0.16

 hatta

0.14

ijn

0.14

 importantly

0.14

 addAction

0.13

 Wander

0.13

rog

0.13

etc

0.13

 Worlds

0.13

Activations Density 0.117%

conjunctions, particularly the word "and"

No Comments

No Known Activations