INDEX

Explanations

names of places and people, primarily in European contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

idad

-0.15

os

-0.15

antal

-0.14

ê·ł

-0.14

verages

-0.14

ÅĻiv

-0.14

voor

-0.14

Īĺ

-0.14

.Framework

-0.14

essel

-0.13

POSITIVE LOGITS

mann

0.19

Mog

0.16

OI

0.15

isseur

0.14

mans

0.14

ä¿

0.14

ìłľ

0.14

èī¾

0.14

Activations Density 0.432%

names of places and people, primarily in European contexts

No Comments

No Known Activations