INDEX

Explanations

quoted speech and dialogue

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

ften

-0.15

ember

-0.15

fen

-0.14

_FWD

-0.14

 åŁ

-0.14

 Wander

-0.14

PMC

-0.14

Å¼e

-0.14

lander

-0.14

izza

-0.13

POSITIVE LOGITS

 Moreno

0.17

comings

0.16

ardin

0.16

 metav

0.15

redi

0.15

 overt

0.15

-bind

0.15

ross

0.14

ilip

0.14

ToProps

0.14

Activations Density 0.171%

quoted speech and dialogue

No Comments

No Known Activations