INDEX

Explanations

references to local entities or organizations in a community or setting

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

istrovstvÃŃ

-0.18

ording

-0.16

Ð²Ð¾Ð»

-0.15

ÌĢ

-0.15

isher

-0.15

adla

-0.14

peria

-0.14

hana

-0.14

Rag

-0.14

-0.13

POSITIVE LOGITS

 whom

0.21

who

0.16

who

0.16

 affected

0.15

 whose

0.15

ort

0.15

Ĭ¶

0.14

ought

0.14

_advance

0.14

Activations Density 0.259%

references to local entities or organizations in a community or setting

No Comments

No Known Activations