INDEX

Explanations

company names

oai_token-act-pair · gpt-3.5-turbo

references to companies in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

alam

-0.79

bian

-0.69

eful

-0.68

cue

-0.67

ilib

-0.67

lyn

-0.67

loe

-0.66

itives

-0.66

iggins

-0.65

icted

-0.64

POSITIVE LOGITS

 executives

1.11

 headquartered

1.08

wide

0.94

 specializing

0.94

 manufactures

0.91

 CEOs

0.90

 founders

0.90

 subsidiary

0.90

 logos

0.86

CEO

0.84

Activations Density 0.077%

company names

references to companies in the text

No Comments

No Known Activations