INDEX

Explanations

phrases indicating one-on-one interactions or meetings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

enstein

-0.17

angkan

-0.16

ãĥ¼ãĥĭ

-0.16

enga

-0.15

kowski

-0.15

estate

-0.15

nt

-0.14

éĸĵ

-0.14

conte

-0.14

ifornia

-0.14

POSITIVE LOGITS

zig

0.16

jed

0.15

à¸¥à¸²à¸Ķ

0.15

 withObject

0.14

 Ridley

0.14

ÑĢÑĸÐ±

0.14

CHA

0.14

cha

0.14

 VÅ¡ech

0.14

icularly

0.13

Activations Density 0.010%

phrases indicating one-on-one interactions or meetings

No Comments

No Known Activations