INDEX

Explanations

words and phrases related to tea

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

 hala

-0.14

tom

-0.14

crt

-0.14

.getJSONObject

-0.14

ottage

-0.13

 butcher

-0.13

isko

-0.13

jon

-0.13

iver

-0.13

argent

-0.13

POSITIVE LOGITS

tea

0.64

Tea

0.57

 teas

0.57

tea

0.52

Te

0.50

èĮ¶

0.47

-te

0.46

Te

0.44

te

0.44

TE

0.42

Activations Density 0.143%

words and phrases related to tea

No Comments

No Known Activations