INDEX

Explanations

terms related to negotiations and agreements between entities, especially involving government bodies

oai_token-act-pair · gpt-3.5-turbo

mentions of governmental and authoritative entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

javascript

-0.64

rb

-0.62

ãĥ¼ãĥĨãĤ£

-0.62

ilyn

-0.60

lopp

-0.60

ãĥ£

-0.60

Wr

-0.59

 Pengu

-0.59

resy

-0.58

 lives

-0.57

POSITIVE LOGITS

 regarding

0.90

 whereby

0.78

 requesting

0.77

 concerning

0.77

 guaranteeing

0.75

via

0.74

 alleging

0.74

 allowing

0.72

izo

0.71

 supplying

0.67

Activations Density 0.445%

terms related to negotiations and agreements between entities, especially involving government bodies

mentions of governmental and authoritative entities

No Comments

No Known Activations