INDEX

Explanations

words related to news and information

oai_token-act-pair · gpt-3.5-turbo

frequently occurring phrases or clauses that provide additional information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

inar

-0.87

minster

-0.77

ema

-0.76

eman

-0.70

 intermediary

-0.68

robe

-0.67

inated

-0.67

anchester

-0.67

emaker

-0.66

oval

-0.65

POSITIVE LOGITS

 however

0.73

"[

0.71

there

0.68

"(

0.68

dos

0.67

 although

0.66

 quoting

0.66

 namely

0.66

we

0.65

"'

0.64

Activations Density 0.180%

words related to news and information

frequently occurring phrases or clauses that provide additional information

No Comments

No Known Activations