INDEX

Explanations

places or specific locations

oai_token-act-pair · gpt-3.5-turbo

phrases indicating location or presence in a specific place

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

oldown

-0.72

tyr

-0.70

 behav

-0.69

 sshd

-0.69

 fortun

-0.68

 withdrawn

-0.66

*/(

-0.65

NetMessage

-0.64

<?

-0.64

POSITIVE LOGITS

abase

0.67

swick

0.67

 Southeast

0.64

 Pathfinder

0.64

 Chad

0.64

soType

0.63

Hal

0.62

 Cathedral

0.62

 Cannes

0.61

 Arizona

0.60

Activations Density 0.080%

places or specific locations

phrases indicating location or presence in a specific place

No Comments

No Known Activations