INDEX

Explanations

phrases indicating something is too much or too intense for a particular situation or individual

oai_token-act-pair · gpt-3.5-turbo

phrases indicating limitations or excessiveness

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

MORE

-0.72

nex

-0.72

 Accessed

-0.70

along

-0.70

iece

-0.69

chenko

-0.68

named

-0.67

ipple

-0.67

onym

-0.66

ener

-0.65

POSITIVE LOGITS

 purposes

0.99

 beginners

0.94

geries

0.94

 anything

0.83

 comparison

0.81

 everyday

0.79

 anybody

0.78

 awhile

0.77

me

0.77

 anyone

0.75

Activations Density 0.195%

phrases indicating something is too much or too intense for a particular situation or individual

phrases indicating limitations or excessiveness

No Comments

No Known Activations