INDEX

Explanations

phrases related to specific places or destinations

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate a type of destination or purpose

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

notations

-0.78

htar

-0.68

 cords

-0.67

dad

-0.64

 Transcript

-0.63

urances

-0.61

 warranties

-0.61

 curves

-0.61

ptive

-0.60

Å¡

-0.60

POSITIVE LOGITS

 sorts

0.78

 beginners

0.77

inel

0.71

geries

0.71

 disguise

0.69

 behold

0.69

 everyday

0.67

 beginner

0.66

 many

0.66

 distraction

0.66

Activations Density 0.264%

phrases related to specific places or destinations

phrases that indicate a type of destination or purpose

No Comments

No Known Activations