INDEX

Explanations

words related to place names

oai_token-act-pair · gpt-3.5-turbo

elements related to storytelling or folklore

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

 carbohyd

-0.72

 millenn

-0.69

 suspic

-0.64

 corrid

-0.64

PDATE

-0.63

 referen

-0.63

ccording

-0.62

 skelet

-0.61

 tremend

-0.60

 gobl

-0.59

POSITIVE LOGITS

phia

0.70

idae

0.69

 Mountains

0.69

Â®

0.69

 Valley

0.66

 Gardens

0.66

ai

0.63

 Heights

0.63

 Manor

0.63

 Island

0.62

Activations Density 0.443%

words related to place names

elements related to storytelling or folklore

No Comments

No Known Activations