INDEX

Explanations

the prefix "ultim-" at various degrees of activation

oai_token-act-pair · gpt-3.5-turbo

instances of the suffix "im" indicating the prefix of various words

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

lace

-0.67

ĸļ

-0.67

DERR

-0.65

hurst

-0.64

chel

-0.64

Lon

-0.62

 Holland

-0.61

 Houses

-0.61

grass

-0.60

pard

-0.59

POSITIVE LOGITS

onial

1.17

ization

1.09

isations

1.08

izations

1.08

izes

1.08

izing

1.06

ized

1.05

ises

0.99

etric

0.97

ulation

0.95

Activations Density 0.070%

the prefix "ultim-" at various degrees of activation

instances of the suffix "im" indicating the prefix of various words

No Comments

No Known Activations