INDEX

Explanations

names or partial names of individuals

oai_token-act-pair · gpt-3.5-turbo

specific names or identifiers

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 bureaucr

-0.77

 corridors

-0.76

 acron

-0.72

TRA

-0.72

 prost

-0.70

 shoulder

-0.69

 multim

-0.68

 rift

-0.67

 rain

-0.67

HOT

-0.66

POSITIVE LOGITS

ney

1.03

rex

0.99

eware

0.99

sted

0.99

ault

0.99

atron

0.98

ember

0.98

yon

0.98

artz

0.97

bach

0.97

Activations Density 0.157%

names or partial names of individuals

specific names or identifiers

No Comments

No Known Activations