INDEX

Explanations

file and image-related terminology

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hra

-0.17

kola

-0.17

ilden

-0.16

/Dk

-0.15

roe

-0.14

 Econ

-0.14

MN

-0.14

 ConnectionState

-0.14

ennie

-0.14

POSITIVE LOGITS

.servlet

0.15

obot

0.14

/rfc

0.14

_qs

0.14

rut

0.14

UGINS

0.14

OUNDS

0.14

ecs

0.14

yal

0.14

RFC

0.13

Activations Density 0.018%

file and image-related terminology

No Comments

No Known Activations