INDEX

Explanations

words related to debris, destruction, and accidents

oai_token-act-pair · gpt-3.5-turbo

terms related to debris and destruction

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

vern

-0.80

conn

-0.70

erity

-0.70

Pis

-0.68

pher

-0.68

ebus

-0.67

Revolution

-0.67

zee

-0.66

Adapt

-0.66

ois

-0.65

POSITIVE LOGITS

 debris

1.20

shed

0.94

 piles

0.92

 fragments

0.90

 particles

0.88

 wreckage

0.82

 heap

0.82

 remnants

0.82

 dumps

0.81

 pile

0.77

Activations Density 0.013%

words related to debris, destruction, and accidents

terms related to debris and destruction

No Comments

No Known Activations