INDEX

Explanations

key terms related to actions, states, or resources, particularly those indicating necessity or presence

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 extrapol

-0.15

rido

-0.15

obo

-0.14

uff

-0.14

opers

-0.14

eurs

-0.14

 smells

-0.13

ofi

-0.13

Ã¡c

-0.13

ella

-0.13

POSITIVE LOGITS

rosse

0.16

evin

0.16

 LENG

0.14

 Mind

0.14

onian

0.14

 Vincent

0.14

marvin

0.14

.embed

0.14

 Bomb

0.14

ombo

0.14

Activations Density 0.009%

key terms related to actions, states, or resources, particularly those indicating necessity or presence

No Comments

No Known Activations