INDEX

Explanations

references to keys and values in data structures

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

ep

-0.14

court

-0.14

aking

-0.14

Ð°ÑĢÐ°Ð½

-0.14

egin

-0.14

/utility

-0.14

itia

-0.13

le

-0.13

 Ð´Ð¾Ð»

-0.13

ast

-0.13

POSITIVE LOGITS

Ð²Ð¸ÑĤ

0.15

llib

0.15

ustos

0.15

zers

0.15

ÃŃrk

0.14

erosis

0.14

ãĥ³ãĥĦ

0.14

cplusplus

0.14

 prest

0.14

isle

0.14

Activations Density 0.200%

references to keys and values in data structures

No Comments

No Known Activations