INDEX

Explanations

code-related structures and functions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

eter

-0.15

 rusty

-0.14

ach

-0.14

yp

-0.14

uman

-0.14

 scratch

-0.14

 Dani

-0.14

 Uncategorized

-0.13

Gir

-0.13

 getArguments

-0.13

POSITIVE LOGITS

astle

0.17

.wrap

0.16

eways

0.16

.createObject

0.15

marshall

0.14

ÎµÎ²

0.14

<decltype

0.14

ãĥĥ

0.14

{'

0.14

Î¿ÏĤ

0.14

Activations Density 0.175%

code-related structures and functions

No Comments

No Known Activations