INDEX

Explanations

words related to wood and materials made of wood

oai_token-act-pair · gpt-3.5-turbo

references to materials, particularly wood and glass

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

 volunt

-0.71

zee

-0.70

agara

-0.69

iqueness

-0.68

bably

-0.66

nesota

-0.65

ipeg

-0.64

 therapist

-0.64

 Teachers

-0.64

ibling

-0.63

POSITIVE LOGITS

 resin

0.77

agall

0.74

 quar

0.71

 artific

0.71

thur

0.70

 encl

0.69

 slic

0.68

idon

0.67

odon

0.66

neck

0.66

Activations Density 0.514%

words related to wood and materials made of wood

references to materials, particularly wood and glass

No Comments

No Known Activations