INDEX

Explanations

information related to specific events or announcements

oai_token-act-pair · gpt-3.5-turbo

constructions that contain the verb "is" to signal states or descriptions

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

inav

-0.72

 summers

-0.68

 stake

-0.67

know

-0.66

 cultures

-0.65

bury

-0.64

 Isles

-0.63

 occupations

-0.63

ossom

-0.62

 throats

-0.61

POSITIVE LOGITS

 comprised

1.15

 reminiscent

1.09

 slated

1.06

 intended

1.04

 aimed

1.03

 supposed

1.01

 titled

1.01

 designed

1.01

 geared

1.00

 meant

0.97

Activations Density 0.302%

information related to specific events or announcements

constructions that contain the verb "is" to signal states or descriptions

No Comments

No Known Activations