INDEX

Explanations

names or terms related to a specific individual

oai_token-act-pair · gpt-3.5-turbo

the mention of specific names related to an individual involved in significant events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

rified

-0.74

mitting

-0.70

mort

-0.69

conscious

-0.69

ienced

-0.68

aminer

-0.68

filled

-0.67

 tremend

-0.67

lasses

-0.66

cribed

-0.66

POSITIVE LOGITS

Äĩ

0.92

yah

0.88

ason

0.82

ya

0.82

cki

0.82

asing

0.80

zon

0.77

Äį

0.77

ensis

0.74

ela

0.74

Activations Density 0.011%

names or terms related to a specific individual

the mention of specific names related to an individual involved in significant events

No Comments

No Known Activations