INDEX

Explanations

words related to specific locations or proper nouns, specifically "Zilker" in this case

oai_token-act-pair · gpt-3.5-turbo

occurrences of names or terms associated with specific locations or organizations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 humble

-0.63

ãĤ¦ãĤ¹

-0.61

Interstitial

-0.61

 conditioned

-0.60

 untold

-0.60

Spoiler

-0.60

 strained

-0.58

 faint

-0.57

taboola

-0.56

IRD

-0.56

POSITIVE LOGITS

anski

0.88

illi

0.83

itsch

0.79

enger

0.78

geist

0.76

urgy

0.73

heng

0.68

zag

0.68

otti

0.67

abyte

0.67

Activations Density 0.089%

words related to specific locations or proper nouns, specifically "Zilker" in this case

occurrences of names or terms associated with specific locations or organizations

No Comments

No Known Activations