INDEX

Explanations

names or terms related to specific locations and individuals

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names of places or people

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

lda

-0.78

 Traff

-0.70

umerable

-0.64

;;;;;;;;;;;;

-0.64

 Violence

-0.64

lder

-0.63

 Rwanda

-0.63

ORGE

-0.63

çīĪ

-0.61

selage

-0.61

POSITIVE LOGITS

arine

1.06

idge

1.04

inet

0.99

aband

0.97

atche

0.94

sburgh

0.92

ards

0.92

arians

0.90

ees

0.89

arding

0.89

Activations Density 0.054%

names or terms related to specific locations and individuals

proper nouns, particularly names of places or people

No Comments

No Known Activations