INDEX

Explanations

references to projects and programming concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

Disclosure

-0.14

nech

-0.14

arget

-0.14

_locals

-0.13

orelease

-0.13

 Frames

-0.13

Ð¯Ðº

-0.13

 species

-0.13

 ÏĢÎ±ÏģÎ¬

-0.13

plorer

-0.13

POSITIVE LOGITS

0.15

imd

0.15

zych

0.15

ê·ł

0.14

aily

0.14

ores

0.14

oyer

0.14

 defaultManager

0.14

NSS

0.13

Activations Density 0.837%

references to projects and programming concepts

No Comments

No Known Activations