INDEX

Explanations

phrases related to various types of damage

oai_token-act-pair · gpt-3.5-turbo

terms related to damage and its consequences

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

KER

-0.73

zee

-0.71

zsche

-0.67

 Reloaded

-0.67

soDeliveryDate

-0.65

_>

-0.62

 foss

-0.62

uchs

-0.61

zl

-0.60

ullivan

-0.60

POSITIVE LOGITS

 inflicted

1.33

 wrought

1.11

 incurred

0.96

 done

0.93

 mitigation

0.93

 damage

0.92

 sustained

0.91

damage

0.90

 caused

0.90

 dealt

0.82

Activations Density 0.071%

phrases related to various types of damage

terms related to damage and its consequences

No Comments

No Known Activations