INDEX

Explanations

occurrences of the word "the"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

yu

-0.13

	StringBuffer

-0.13

 duÅ¼

-0.12

_pemb

-0.12

undy

-0.12

,:,

-0.12

edin

-0.12

kola

-0.12

rint

-0.12

_ENABLE

-0.12

POSITIVE LOGITS

 same

0.84

same

0.72

 SAME

0.65

 Same

0.60

Same

0.57

.same

0.57

_same

0.55

sam

0.54

SAME

0.54

 sama

0.53

Activations Density 0.075%

occurrences of the word "the"

No Comments

No Known Activations