INDEX

Explanations

phrases related to locations or places, specifically emphasizing the word "ant."

oai_token-act-pair · gpt-3.5-turbo

the word "ant" 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 @ 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

UTION

-0.87

çīĪ

-0.77

STON

-0.74

earchers

-0.70

TOR

-0.68

displayText

-0.67

MET

-0.67

inventoryQuantity

-0.65

Â¯Â¯Â¯Â¯Â¯Â¯Â¯Â¯

-0.64

ãĤ¡

-0.64

POSITIVE LOGITS

ucket

1.04

anamo

1.02

heon

1.00

ioch

0.95

asma

0.92

ropy

0.91

asia

0.87

agonist

0.87

rell

0.87

ieri

0.86

Activations Density 0.051%

phrases related to locations or places, specifically emphasizing the word "ant."

the word "ant" in various contexts

No Comments

No Known Activations