INDEX

Explanations

terms related to war or conflict situations

oai_token-act-pair · gpt-3.5-turbo

terms related to military equipment or structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

âĸ¬âĸ¬

-0.73

ollar

-0.70

âĸ¬

-0.68

Bloom

-0.67

è¦ļéĨĴ

-0.67

 Masquerade

-0.66

Trend

-0.62

draft

-0.62

aternity

-0.62

CENT

-0.62

POSITIVE LOGITS

imus

0.77

ĪĴ

0.71

wo

0.65

ams

0.63

 routines

0.62

 selector

0.60

 programmed

0.60

 rigged

0.59

ved

0.59

 liner

0.58

Activations Density 0.425%

terms related to war or conflict situations

terms related to military equipment or structures

No Comments

No Known Activations