INDEX

Explanations

references to individuals, particularly names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 19-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L19R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.19.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

erokee

-0.15

åģ

-0.15

inator

-0.14

podob

-0.14

âĸ²

-0.14

.ReadFile

-0.14

Ð°ÑĢÑĩ

-0.14

STANCE

-0.14

rai

-0.14

EVT

-0.14

POSITIVE LOGITS

 Chris

0.17

umbnails

0.16

uma

0.14

hyp

0.14

Pac

0.14

 Christopher

0.14

 bull

0.13

 plein

0.13

 Wilson

0.13

 Phil

0.13

Activations Density 0.015%

references to individuals, particularly names

No Comments

No Known Activations