INDEX

Explanations

nested object and array structures in code

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

rug

-0.15

;base

-0.14

atin

-0.14

ENDOR

-0.14

pron

-0.13

à¸«à¸²à¸£

-0.13

 Oaks

-0.13

áº½

-0.13

ipur

-0.13

bon

-0.13

POSITIVE LOGITS

ÃŃl

0.16

lew

0.15

ãĥĥãĥĪ

0.15

imas

0.14

ts

0.14

 deferred

0.14

YLE

0.14

inely

0.14

.infinity

0.14

Ã¡ky

0.13

Activations Density 0.032%

nested object and array structures in code

No Known Activations