INDEX

Explanations

structural elements and formatting indicators in code snippets

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

-0.18

hil

-0.17

Ãªn

-0.16

 resorts

-0.14

á»©t

-0.14

jan

-0.14

agne

-0.14

enis

-0.14

lo

-0.14

ensen

-0.14

POSITIVE LOGITS

lemn

0.16

?}",

0.15

pickle

0.14

overy

0.14

 tender

0.14

.failure

0.14

ATIC

0.14

EIF

0.14

upy

0.14

ì£¼ìĿĺ

0.14

Activations Density 0.078%

structural elements and formatting indicators in code snippets

No Comments

No Known Activations