INDEX

Explanations

references to the word "ant" and its variations, indicating a specific focus on that term in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

rl

-0.17

rig

-0.17

rint

-0.16

ra

-0.16

strup

-0.16

ryo

-0.15

hib

-0.15

ront

-0.15

ract

-0.15

riz

-0.15

POSITIVE LOGITS

0.23

ucket

0.22

yne

0.22

elope

0.20

ing

0.20

enna

0.19

woord

0.19

yre

0.19

werp

0.18

ech

0.17

Activations Density 0.033%

references to the word "ant" and its variations, indicating a specific focus on that term in different contexts

No Comments

No Known Activations