INDEX

Explanations

phrases related to health and safety incidents or situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

ivot

-0.16

okie

-0.15

ifest

-0.15

icone

-0.14

ooks

-0.14

ocus

-0.14

èĤĸ

-0.14

 fork

-0.14

ifs

-0.13

POSITIVE LOGITS

»

0.15

thal

0.14

Kov

0.14

ngth

0.14

POSITORY

0.14

ungs

0.13

 ë§

0.13

 datap

0.13

 neighbourhood

0.13

Activations Density 0.021%

phrases related to health and safety incidents or situations

No Comments

No Known Activations