INDEX

Explanations

words associated with evaluation and analysis

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

 Plus

-0.15

udd

-0.15

iggins

-0.15

icularly

-0.14

ï»¿#

-0.14

ä¹ĥ

-0.14

 emerg

-0.14

otten

-0.14

Via

-0.13

foon

-0.13

POSITIVE LOGITS

hetto

0.16

tainment

0.15

 ÐºÐ¸ÑĢ

0.14

 Leban

0.14

eneric

0.14

prech

0.14

lest

0.13

ocale

0.13

loth

0.13

utan

0.13

Activations Density 0.513%

words associated with evaluation and analysis

No Comments

No Known Activations