INDEX

Explanations

references to specific geographical locations, particularly the city of Avon

oai_token-act-pair · gpt-3.5-turbo

words and names associated with identifiers related to the fictional character "Avon."

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

 quarters

-0.70

 Democr

-0.67

itsch

-0.66

ocene

-0.61

ioch

-0.60

telling

-0.60

liners

-0.57

ricting

-0.56

ocr

-0.56

 axes

-0.55

POSITIVE LOGITS

AIDS

0.67

ãĥĨãĤ£

0.67

role

0.64

icious

0.63

CLASSIFIED

0.62

ILE

0.62

assis

0.61

dor

0.60

orbit

0.60

urate

0.60

Activations Density 0.090%

references to specific geographical locations, particularly the city of Avon

words and names associated with identifiers related to the fictional character "Avon."

No Comments

No Known Activations