INDEX

Explanations

terms related to explosive devices and actions

oai_token-act-pair · gpt-3.5-turbo

terms related to explosives and detonations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

esan

-0.73

Sea

-0.71

aird

-0.71

TB

-0.70

SEA

-0.69

croft

-0.67

 Acqu

-0.67

anson

-0.67

ACC

-0.66

bm

-0.66

POSITIVE LOGITS

 detonated

1.34

 deton

1.31

 explosives

0.88

 bomb

0.86

 bombs

0.85

 deterrent

0.83

ashtra

0.83

 bombing

0.81

 rall

0.81

 bombings

0.80

Activations Density 0.009%

terms related to explosive devices and actions

terms related to explosives and detonations

No Comments

No Known Activations