INDEX

Explanations

references to geographic locations and intersections in navigation contexts

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

insky

-0.15

antz

-0.15

esel

-0.14

held

-0.14

Äįe

-0.14

ighth

-0.14

 held

-0.14

Äįi

-0.14

toast

-0.14

imity

-0.14

POSITIVE LOGITS

inize

0.16

 junction

0.15

plits

0.15

yen

0.15

yx

0.14

_UNUSED

0.14

junction

0.14

orm

0.14

 Leisure

0.14

quals

0.14

Activations Density 0.078%

references to geographic locations and intersections in navigation contexts

No Comments

No Known Activations