INDEX

Explanations

symbols or markers indicating sections or transformations in content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

 Bols

-0.18

ÛĮÙĩ

-0.16

utzt

-0.15

aget

-0.15

fol

-0.15

 Brennan

-0.15

 KrÃ¡l

-0.14

ÑĢÐ°Ð±

-0.14

.LatLng

-0.14

§

-0.13

POSITIVE LOGITS

ullet

0.16

ovky

0.15

thora

0.15

å¾®è½¯éĽħé»ĳ

0.15

igos

0.15

Sb

0.14

 chains

0.14

 tslint

0.14

Ð³Ð¾ÑĤÐ¾Ð²

0.14

lim

0.14

Activations Density 0.005%

symbols or markers indicating sections or transformations in content

No Comments

No Known Activations