INDEX

Explanations

words associated with specific individuals or 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_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

pora

-0.66

 Skydragon

-0.65

 actionGroup

-0.62

 Cortex

-0.62

 quarters

-0.61

 "$:/

-0.61

 wagon

-0.60

NCT

-0.59

 synaptic

-0.59

 membr

-0.59

POSITIVE LOGITS

rill

0.96

iland

0.95

inki

0.77

lein

0.76

chery

0.75

vity

0.74

wear

0.74

glers

0.73

zinski

0.73

okia

0.71

Activations Density 0.020%

words associated with specific individuals or entities

No Comments

No Known Activations