INDEX

Explanations

references to specific political administrations, particularly focusing on actions or decisions

oai_token-act-pair · gpt-3.5-turbo

references to a specific government administration

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

actionDate

-0.77

Ô

-0.71

 ILCS

-0.71

aceous

-0.69

#$#$

-0.66

ãĤ¡

-0.65

vill

-0.64

 Plate

-0.64

Kin

-0.63

Ward

-0.63

POSITIVE LOGITS

 secretaries

0.86

 officials

0.84

 appoint

0.79

 sanctioned

0.78

 secretary

0.76

 tasked

0.75

 bureaucracy

0.74

ahime

0.73

 playbook

0.73

 shutdown

0.72

Activations Density 0.025%

references to specific political administrations, particularly focusing on actions or decisions

references to a specific government administration

No Comments

No Known Activations