INDEX

Explanations

programming-related keywords and methods that indicate actions or operations being performed

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

leaning

-0.16

phia

-0.16

oulouse

-0.15

leans

-0.15

icons

-0.14

arg

-0.14

aille

-0.14

PDOException

-0.14

endum

-0.13

IG

-0.13

POSITIVE LOGITS

 Morr

0.14

ØŃÙĨ

0.14

isti

0.14

psc

0.14

_bug

0.14

ï»¿using

0.14

 anale

0.14

/System

0.13

/application

0.13

aska

0.13

Activations Density 0.025%

programming-related keywords and methods that indicate actions or operations being performed

No Comments

No Known Activations