INDEX

Explanations

repeated structures or patterns in code, particularly related to conditions and the end of statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

linger

-0.15

 conven

-0.15

amus

-0.15

avec

-0.15

vary

-0.14

ech

-0.14

 Damen

-0.14

JSONObject

-0.14

Vie

-0.14

vx

-0.14

POSITIVE LOGITS

egan

0.17

ï»¿↵↵

0.15

_subscribe

0.15

SystemService

0.15

ë¡ľëĵľ

0.15

wind

0.14

 Cheng

0.14

Subset

0.14

Byl

0.13

opes

0.13

Activations Density 0.222%

repeated structures or patterns in code, particularly related to conditions and the end of statements

No Comments

No Known Activations