INDEX

Explanations

words related to traps

oai_token-act-pair · gpt-3.5-turbo

the concept of traps 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 @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Merit

-0.71

edly

-0.70

 Kardashian

-0.66

league

-0.66

issance

-0.63

played

-0.62

 normalized

-0.61

 Glory

-0.60

sis

-0.60

 concerned

-0.60

POSITIVE LOGITS

door

1.35

trap

1.33

 trap

1.23

 traps

1.13

doors

1.08

 Trap

1.06

 Traps

0.99

 trapping

0.87

idon

0.86

 Hole

0.83

Activations Density 0.013%

words related to traps

the concept of traps in various contexts

No Comments

No Known Activations