INDEX

Explanations

words related to specific locations, potentially referring to places or events

oai_token-act-pair · gpt-3.5-turbo

repeated occurrences of the substring "ri"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

é¾

-0.84

dayName

-0.82

KNOWN

-0.66

 horizon

-0.64

 Fourth

-0.63

 chips

-0.62

 Hilbert

-0.62

 Suffolk

-0.61

lot

-0.61

CPC

-0.61

POSITIVE LOGITS

ving

1.14

vals

1.13

quez

1.11

jen

1.07

zona

1.04

Ã«

1.01

ja

1.00

jah

0.96

zzle

0.96

quet

0.96

Activations Density 0.009%

words related to specific locations, potentially referring to places or events

repeated occurrences of the substring "ri"

No Comments

No Known Activations