INDEX

Explanations

phrases containing the word "said" followed by additional information

oai_token-act-pair · gpt-3.5-turbo

occurrences of the single quotation mark character

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

 Hendricks

-0.70

 printer

-0.70

 Archdemon

-0.69

nown

-0.68

 printers

-0.67

 lapse

-0.65

 Izan

-0.65

 Mechdragon

-0.64

isphere

-0.64

 Leilan

-0.62

POSITIVE LOGITS

cause

1.30

til

1.25

Interstitial

1.14

Mech

1.11

tis

1.11

taboola

1.08

-'

1.04

round

1.00

',

0.92

keeper

0.88

Activations Density 0.041%

phrases containing the word "said" followed by additional information

occurrences of the single quotation mark character

No Comments

No Known Activations