INDEX

Explanations

words related to things that are undeveloped or not fully realized

oai_token-act-pair · gpt-3.5-turbo

terms related to undeveloped or underprivileged conditions

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

antha

-0.80

ingham

-0.79

hetti

-0.77

lar

-0.73

DOC

-0.73

é¾įå¥ĳå£«

-0.69

 Heads

-0.68

 GEAR

-0.67

 Cosponsors

-0.67

osterone

-0.65

POSITIVE LOGITS

velop

1.00

merce

0.92

 unde

0.85

FINE

0.84

lete

0.83

chnology

0.83

jection

0.82

ploy

0.77

clave

0.77

eteen

0.76

Activations Density 0.054%

words related to things that are undeveloped or not fully realized

terms related to undeveloped or underprivileged conditions

No Comments

No Known Activations