INDEX

Explanations

names of authors and their affiliations or citations within academic contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

itag

-0.16

boru

-0.15

tw

-0.15

iri

-0.15

oku

-0.15

orre

-0.14

 association

-0.14

Ung

-0.14

iate

-0.14

AMA

-0.14

POSITIVE LOGITS

haus

0.16

edImage

0.16

harma

0.15

 FindObjectOfType

0.14

insula

0.14

.middleware

0.13

ertest

0.13

ClassLoader

0.13

jr

0.13

_FULLSCREEN

0.13

Activations Density 0.001%

names of authors and their affiliations or citations within academic contexts

No Comments

No Known Activations