INDEX

Explanations

repeated punctuation marks, particularly commas, within a structured narrative or dataset context

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

bish

-0.15

 Ð½Ð°ÑĤÑĥ

-0.14

ogle

-0.14

icken

-0.13

mnt

-0.13

WARDED

-0.13

baugh

-0.13

latin

-0.13

 extern

-0.13

 Ð¾Ñĩ

-0.13

POSITIVE LOGITS

Criterion

0.16

uptime

0.16

LIK

0.15

quia

0.15

PointerType

0.14

Ã©ra

0.13

ako

0.13

å©Ĩ

0.13

bette

0.13

Sizer

0.13

Activations Density 0.002%

repeated punctuation marks, particularly commas, within a structured narrative or dataset context

No Comments

No Known Activations