INDEX

Explanations

positive descriptors or assessments

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

ordan

-0.14

halten

-0.14

Fon

-0.14

adiens

-0.14

rote

-0.14

isu

-0.13

hl

-0.13

iah

-0.13

urs

-0.13

POSITIVE LOGITS

-tm

0.16

åĿ¡

0.15

.quest

0.15

Äįer

0.14

 coincidence

0.14

matter

0.14

 matter

0.13

 STEP

0.13

ching

0.13

ignon

0.13

Activations Density 0.061%

positive descriptors or assessments

No Comments

No Known Activations