INDEX

Explanations

terms related to proper names or brands

oai_token-act-pair · gpt-3.5-turbo

proper nouns and brand names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

internal

-1.07

arching

-1.00

tc

-1.00

unc

-1.00

adjust

-0.97

open

-0.94

take

-0.93

overe

-0.93

esp

-0.92

exec

-0.92

POSITIVE LOGITS

 Dome

1.02

 Butt

0.90

 Suit

0.90

 Ones

0.88

Gul

0.88

 Shack

0.86

 Widow

0.84

Man

0.84

 Friends

0.84

 Skull

0.84

Activations Density 0.282%

terms related to proper names or brands

proper nouns and brand names

No Comments

No Known Activations