INDEX

Explanations

references to geographical features, particularly hills and mountains

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

exus

-0.16

asta

-0.16

ayer

-0.16

taj

-0.16

strup

-0.15

Ã³st

-0.15

istant

-0.14

 Mile

-0.14

ediator

-0.14

iero

-0.14

POSITIVE LOGITS

pong

0.16

ward

0.15

0.14

icion

0.14

_DAC

0.14

owy

0.14

 infertility

0.14

WARDED

0.13

arding

0.13

.setTitle

0.13

Activations Density 0.072%

references to geographical features, particularly hills and mountains

No Comments

No Known Activations