INDEX

Explanations

various data points, particularly numbers and their associated contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

elho

-0.19

usra

-0.17

habi

-0.17

 ^{Â°}

-0.15

afari

-0.15

hab

-0.15

URRE

-0.15

 Kenn

-0.14

urette

-0.14

ROTO

-0.14

POSITIVE LOGITS

ÑģÐ»Ð¾Ð²

0.16

ha

0.16

ÏĦÎ¬

0.15

Jer

0.15

ordinate

0.15

Jar

0.15

jar

0.14

éħį

0.14

lear

0.14

Activations Density 0.002%

various data points, particularly numbers and their associated contexts

No Comments

No Known Activations