INDEX

Explanations

phrases introducing information or arguments

oai_token-act-pair · gpt-3.5-turbo

occurrences of phrases starting with "As"

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

thro

-0.75

eeee

-0.67

utical

-0.66

DOWN

-0.65

ALLY

-0.63

paralleled

-0.63

apons

-0.62

||||

-0.61

osc

-0.60

ãģł

-0.60

POSITIVE LOGITS

bestos

1.20

pects

1.11

king

1.11

piring

1.09

semb

1.09

ymm

1.06

phalt

1.04

 evidenced

1.01

piration

1.00

semble

0.98

Activations Density 0.086%

phrases introducing information or arguments

occurrences of phrases starting with "As"

No Comments

No Known Activations