INDEX

Explanations

websites or online resources

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "info" or variations thereof

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ŃĶ

-0.73

©¶æ

-0.70

sticks

-0.69

lock

-0.68

locks

-0.68

pher

-0.67

stick

-0.65

senal

-0.63

amping

-0.63

 notor

-0.63

POSITIVE LOGITS

llor

0.95

useum

0.87

erences

0.85

etta

0.85

asure

0.81

uration

0.81

ablishment

0.80

illery

0.80

tenance

0.79

ession

0.77

Activations Density 0.026%

websites or online resources

occurrences of the term "info" or variations thereof

No Comments

No Known Activations