INDEX

Explanations

references to military organizations, particularly the Air Force

oai_token-act-pair · gpt-3.5-turbo

mentions of the acronym "AF" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Kessler

-0.76

 Gors

-0.73

igators

-0.72

 Blackwell

-0.70

 Muller

-0.70

 Boone

-0.69

 Heller

-0.66

MLG

-0.66

 Podesta

-0.65

 Gott

-0.65

POSITIVE LOGITS

RIC

1.24

SPA

1.00

ħĭ

1.00

rican

0.98

avorite

0.96

rica

0.96

ornia

0.94

lesh

0.93

amily

0.93

chieve

0.92

Activations Density 0.012%

references to military organizations, particularly the Air Force

mentions of the acronym "AF" in various contexts

No Comments

No Known Activations