INDEX

Explanations

occurrences of words related to the action of placing or positioning something

oai_token-act-pair · gpt-3.5-turbo

references to placement or positioning

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

ãģ®éŃĶ

-0.70

 Annotations

-0.69

heimer

-0.67

LEASE

-0.65

UNE

-0.65

Ô

-0.64

FINEST

-0.62

SOURCE

-0.61

 Chronicle

-0.61

IRO

-0.60

POSITIVE LOGITS

ements

1.18

 plac

1.03

ate

1.02

atos

0.98

atron

0.95

ating

0.94

ently

0.93

orescence

0.90

atable

0.86

ated

0.86

Activations Density 0.016%

occurrences of words related to the action of placing or positioning something

references to placement or positioning

No Comments

No Known Activations