INDEX

Explanations

words related to closing or shutting things

oai_token-act-pair · gpt-3.5-turbo

references to closure or endings in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

Magikarp

-0.65

ãĥīãĥ©ãĤ´ãĥ³

-0.61

PET

-0.61

nikov

-0.61

Gry

-0.61

arthy

-0.60

nr

-0.59

chenko

-0.59

 Samoa

-0.58

SPONSORED

-0.58

POSITIVE LOGITS

 doors

1.03

 arteries

0.94

 door

0.85

 throttle

0.84

gap

0.83

 borders

0.82

 curtains

0.81

 pores

0.80

alore

0.80

 reins

0.79

Activations Density 0.086%

words related to closing or shutting things

references to closure or endings in various contexts

No Comments

No Known Activations