INDEX

Explanations

terms related to governmental actions and policies

oai_token-act-pair · gpt-3.5-turbo

discussions surrounding political actions and decisions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.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

 lifes

-0.96

ĸļ

-0.88

 commissions

-0.78

acc

-0.77

 casc

-0.77

 dolphin

-0.77

 indoctr

-0.77

swe

-0.76

 ingest

-0.76

 endeav

-0.75

POSITIVE LOGITS

Either

1.62

Meanwhile

1.60

Assuming

1.58

Sources

1.53

Regardless

1.53

Nevertheless

1.47

Asked

1.45

According

1.42

Should

1.41

Speaking

1.40

Activations Density 0.449%

terms related to governmental actions and policies

discussions surrounding political actions and decisions

No Comments

No Known Activations