INDEX

Explanations

references to the word "Roll" in various contexts

oai_token-act-pair · gpt-3.5-turbo

references to "Roll Call" and related voting procedures

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

ãĥ¤

-0.91

ILE

-0.80

ãĤ»

-0.78

ãĥŁ

-0.72

ãĤ±

-0.71

bub

-0.71

ãĥģ

-0.70

ilities

-0.69

itably

-0.67

 Gemini

-0.66

POSITIVE LOGITS

sheet

0.90

aways

0.86

Roll

0.86

ers

0.85

roll

0.84

outs

0.80

top

0.79

away

0.78

anut

0.78

icking

0.77

Activations Density 0.019%

references to the word "Roll" in various contexts

references to "Roll Call" and related voting procedures

No Comments

No Known Activations