INDEX

Explanations

words that indicate frequency or commonality in actions or descriptions

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

tuk

-0.16

ptest

-0.16

ifax

-0.15

elight

-0.15

kt

-0.15

iginal

-0.15

Pixel

-0.15

Ã³b

-0.15

CJK

-0.14

urette

-0.14

POSITIVE LOGITS

iece

0.17

 Leone

0.15

awai

0.15

likely

0.14

feeding

0.14

runner

0.14

Ø§ÙĦÙī

0.13

 ADVISED

0.13

-match

0.13

 Stanton

0.13

Activations Density 0.042%

words that indicate frequency or commonality in actions or descriptions

No Comments

No Known Activations