INDEX

Explanations

instances of specific terms or phrases that pertain to actions or gameplay in a strategy context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

 oblig

-0.16

emble

-0.16

ello

-0.15

iva

-0.15

illet

-0.14

hcp

-0.14

ucc

-0.14

mixed

-0.14

uss

-0.14

usc

-0.13

POSITIVE LOGITS

peare

0.23

igans

0.17

à¥ģà¤Ĩà¤¤

0.17

/Sh

0.16

inking

0.16

poke

0.15

ÐµÑĢÐ³

0.15

asti

0.14

Ĺi

0.14

Activations Density 0.056%

instances of specific terms or phrases that pertain to actions or gameplay in a strategy context

No Comments

No Known Activations