INDEX

Explanations

numeric values, particularly dates

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

ersiz

-0.18

WER

-0.16

mov

-0.14

amam

-0.14

Ð¾Ð³ÑĢÐ°Ð¼

-0.14

xbe

-0.13

pec

-0.13

 tracks

-0.13

isbury

-0.13

ono

-0.13

POSITIVE LOGITS

 Pleasant

0.15

edException

0.14

@}

0.14

lj

0.14

(skb

0.14

ImGui

0.14

oucher

0.14

ospel

0.13

atsapp

0.13

ØªÙĩØ§

0.13

Activations Density 0.002%

numeric values, particularly dates

No Comments

No Known Activations