INDEX

Explanations

instances where something is specifically used for a particular purpose or activity

oai_token-act-pair · gpt-3.5-turbo

phrases indicating uses or applications of items or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

beat

-0.77

é¾

-0.74

iev

-0.73

ellen

-0.72

uckland

-0.72

ername

-0.70

ouf

-0.68

pan

-0.68

osa

-0.67

umar

-0.67

POSITIVE LOGITS

 purposes

1.45

geries

1.11

 storing

1.10

aging

1.09

 transporting

1.05

 constructing

1.01

gery

1.00

 nefarious

0.95

 verification

0.94

 experimentation

0.94

Activations Density 0.159%

instances where something is specifically used for a particular purpose or activity

phrases indicating uses or applications of items or concepts

No Comments

No Known Activations