INDEX

Explanations

references to the Java programming language and its core classes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

arrant

-0.15

orsk

-0.15

pekt

-0.15

ifice

-0.14

lus

-0.14

ERY

-0.14

lou

-0.14

Ãłng

-0.14

leck

-0.14

POSITIVE LOGITS

gere

0.15

ãĥ³ãĤ¬

0.15

_Global

0.14

 Glob

0.14

/do

0.14

spe

0.13

 _↵↵

0.13

kiye

0.13

Ñı

0.13

Activations Density 0.005%

references to the Java programming language and its core classes

No Comments

No Known Activations