INDEX

Explanations

words related to technical terms or programming syntax

oai_token-act-pair · gpt-3.5-turbo

special characters and symbols in text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

$.

-0.68

 respectively

-0.63

}.

-0.61

]."

-0.61

.).

-0.56

 thereof

-0.55

".

-0.52

%.

-0.51

''.

-0.51

'.

-0.51

POSITIVE LOGITS

 FANTASY

0.45

 âĢº

0.44

iple

0.42

 partName

0.41

 nutshell

0.40

0.38

resa

0.38

reenshots

0.38

:=

0.36

 Posted

0.36

Activations Density 5.555%

words related to technical terms or programming syntax

special characters and symbols in text

No Comments

No Known Activations