INDEX

Explanations

places or concepts that provide safety or support

oai_token-act-pair · gpt-3.5-turbo

concepts related to safety, protection, and support systems

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

ancies

-0.80

qus

-0.78

intent

-0.70

ntax

-0.67

 staggered

-0.65

yss

-0.64

itars

-0.64

©¶æ

-0.64

wagen

-0.64

erity

-0.62

POSITIVE LOGITS

 reminder

0.87

 beacon

0.81

 fodder

0.76

 discipl

0.76

piece

0.76

 anthem

0.75

 entry

0.74

 gateway

0.72

inel

0.71

 magnet

0.70

Activations Density 0.303%

places or concepts that provide safety or support

concepts related to safety, protection, and support systems

No Comments

No Known Activations