INDEX

Explanations

locations or places

oai_token-act-pair · gpt-3.5-turbo

occurrences of specific locations and time-related phrases

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

erity

-0.74

 verbal

-0.72

 rookies

-0.67

 adaptations

-0.65

 jokes

-0.64

 backups

-0.61

 caliber

-0.60

Writ

-0.60

IGHT

-0.60

 assistants

-0.60

POSITIVE LOGITS

fecture

1.12

 suburb

0.98

 suburbs

0.88

 outskirts

0.87

 Borough

0.87

 Cathedral

0.86

 valley

0.85

shire

0.84

 harbour

0.84

 Junction

0.84

Activations Density 0.541%

locations or places

occurrences of specific locations and time-related phrases

No Comments

No Known Activations