INDEX

Explanations

phrases related to conflicts or disputes between entities

oai_token-act-pair · gpt-3.5-turbo

phrases related to conflicts or disputes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

oooooooo

-0.80

ooo

-0.75

+)

-0.72

ãĥ¼ãĥĨãĤ£

-0.72

oooo

-0.70

-|

-0.67

Yeah

-0.66

iva

-0.66

Today

-0.66

orst

-0.65

POSITIVE LOGITS

reach

1.13

tones

1.03

 whether

1.01

 disagreements

0.90

 allegations

0.87

reaching

0.86

 unpaid

0.86

 issues

0.83

 perceived

0.82

 fears

0.82

Activations Density 0.077%

phrases related to conflicts or disputes between entities

phrases related to conflicts or disputes

No Comments

No Known Activations