INDEX

Explanations

phrases and words related to aggressive actions or behaviors

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

ĸļ

-0.98

obyl

-0.90

owship

-0.81

udder

-0.79

psons

-0.78

 Alive

-0.77

ãĤ´ãĥ³

-0.76

artifacts

-0.71

mbuds

-0.70

FORE

-0.70

POSITIVE LOGITS

 aggressive

0.91

 posture

0.83

 tactics

0.82

 behavior

0.81

 toward

0.81

 maneuver

0.80

 stance

0.79

 lobbying

0.78

 aggress

0.77

 maneuvers

0.76

Activations Density 0.018%

phrases and words related to aggressive actions or behaviors

No Comments

No Known Activations