INDEX

Explanations

words related to being deprived or facing challenging circumstances

oai_token-act-pair · gpt-3.5-turbo

adjectives describing negative conditions or states

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

wcsstore

-0.86

BIT

-0.79

atography

-0.79

akes

-0.77

aiden

-0.75

asso

-0.74

quickShipAvailable

-0.74

WithNo

-0.73

azaki

-0.71

LOD

-0.71

POSITIVE LOGITS

 souls

0.83

 youths

0.75

 youngsters

0.71

 veteran

0.71

 former

0.70

 Corinth

0.69

 young

0.69

 masses

0.69

 nerves

0.68

 carc

0.67

Activations Density 0.143%

words related to being deprived or facing challenging circumstances

adjectives describing negative conditions or states

No Comments

No Known Activations