INDEX

Explanations

terms related to unique identifiers and counters in programming contexts

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

ROTO

-0.16

 PROT

-0.16

quals

-0.16

ìĨĮê°ľ

-0.15

-json

-0.15

imore

-0.14

cq

-0.14

POSIT

-0.14

ë»

-0.14

cie

-0.14

POSITIVE LOGITS

izer

0.17

enco

0.17

aler

0.16

ewire

0.15

ä¼¼çļĦ

0.15

OB

0.14

icha

0.14

IFY

0.14

IDO

0.14

Activations Density 0.056%

terms related to unique identifiers and counters in programming contexts

No Comments

No Known Activations