INDEX

Explanations

phrases related to explosive devices

oai_token-act-pair · gpt-3.5-turbo

terms related to explosives and detonators

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

heit

-0.79

esan

-0.78

aird

-0.74

SEA

-0.74

}}}

-0.71

Disney

-0.70

ĻĤ

-0.70

 Borders

-0.70

orses

-0.69

 Shepherd

-0.68

POSITIVE LOGITS

 detonated

1.57

 deton

1.46

 grenades

1.25

 explosives

1.23

 explosions

1.19

 grenade

1.15

 ignition

1.14

 explosion

1.14

 bombs

1.09

 explosive

1.07

Activations Density 0.078%

phrases related to explosive devices

terms related to explosives and detonators

No Comments

No Known Activations