INDEX

Explanations

numbers related to financial indices or events

oai_token-act-pair · gpt-3.5-turbo

references to the "500" in various contexts

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

eworld

-0.82

inion

-0.81

olulu

-0.78

icho

-0.78

worldly

-0.77

colm

-0.77

mort

-0.77

verend

-0.76

odynam

-0.75

plom

-0.74

POSITIVE LOGITS

MHz

0.82

mA

0.82

px

0.79

 watts

0.78

ãĥ¼ãĥĨãĤ£

0.78

Hz

0.77

 Mile

0.77

Hz

0.76

ãĥ©ãĥ³

0.73

ãĤ¼ãĤ¦ãĤ¹

0.72

Activations Density 0.040%

numbers related to financial indices or events

references to the "500" in various contexts

No Comments

No Known Activations