INDEX

Explanations

mentions of the word "communications"

oai_token-act-pair · gpt-3.5-turbo

references to communication activities and technologies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

jury

-0.93

steen

-0.81

gling

-0.80

amina

-0.78

iless

-0.78

endon

-0.76

\/\/

-0.76

Äį

-0.74

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

-0.74

 Naples

-0.72

POSITIVE LOGITS

 communications

0.99

 communication

0.92

communications

0.92

 protocols

0.90

 messages

0.88

 networks

0.86

 channels

0.86

 transmissions

0.84

 theorist

0.81

 protocol

0.80

Activations Density 0.019%

mentions of the word "communications"

references to communication activities and technologies

No Comments

No Known Activations