INDEX

Explanations

technical terminology and code-related constructs

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

sted

-0.17

unal

-0.15

illy

-0.15

NSF

-0.14

 identical

-0.14

ste

-0.13

rex

-0.13

à¸±à¸į

-0.13

ale

-0.13

agal

-0.13

POSITIVE LOGITS

bilt

0.16

essay

0.16

azer

0.15

ynos

0.15

atu

0.15

ãĥ³ãĥĶ

0.14

.hw

0.14

iá»ģm

0.14

inspace

0.14

ÐµÐ·ÑĥÐ»ÑĮÑĤ

0.14

Activations Density 0.127%

technical terminology and code-related constructs

No Comments

No Known Activations