INDEX

Explanations

words related to the term "Ram"

oai_token-act-pair · gpt-3.5-turbo

references to the name "Ram"

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

RAFT

-0.79

 Consent

-0.73

chnology

-0.68

 Bermuda

-0.63

 Dangerous

-0.63

kHz

-0.61

KI

-0.61

 Yankee

-0.60

 Norn

-0.60

ģĸ

-0.60

POSITIVE LOGITS

irez

1.26

rod

1.09

pling

1.09

blers

1.05

pton

1.03

esses

0.98

adan

0.97

ming

0.96

aph

0.94

agna

0.92

Activations Density 0.013%

words related to the term "Ram"

references to the name "Ram"

No Comments

No Known Activations