INDEX

Explanations

specific mentions of titles or organizations, particularly within a structured context

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "the" in various contexts

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

sie

-0.77

gui

-0.68

cial

-0.66

getic

-0.66

âĸł

-0.65

tty

-0.63

iffe

-0.63

illon

-0.62

NetMessage

-0.62

Topics

-0.61

POSITIVE LOGITS

 rest

1.23

 consequ

1.18

 ensuing

1.18

 resultant

1.15

 accompanying

1.13

 surrounding

1.11

 wider

1.04

 attendant

0.99

 subsequent

0.99

 adjoining

0.98

Activations Density 0.182%

specific mentions of titles or organizations, particularly within a structured context

occurrences of the word "the" in various contexts

No Comments

No Known Activations